A New and Improved Whole Life Satisfaction Account of Happiness


Angner, E. (2011). The evolution of eupathics: The historical roots of subjective measures of wellbeing. 

International Journal of Wellbeing, 1(1), 4-41. doi:10.5502/ijw.v1i1.14  

 

Erik Angner 

University of Alabama at Birmingham 

angner@uab.edu 
 

Copyright belongs to the author(s) 

www.internationaljournalofwellbeing.org 

4 

ARTICLE  

 

The evolution of eupathics:  

The historical roots of subjective measures of wellbeing 
 

Erik Angner 

 

 
Abstract: This paper traces the historical roots of subjective measures of wellbeing, that is, 

measures designed to represent happiness, satisfaction, or other ‚positive‛ or desirable mental 

states. While it is often suggested that these measures are a modern invention, I argue that they 

have a long and rich history that conforms to Theodore M. Porter’s general account of 

measurement in social and behavioral science. Subjective measures emerged in marital success 

studies, educational psychology, and personality psychology in the 1920s and 30s, and were 

further shaped by the epidemiology of mental health, gerontology, and the social indicator 

movement in the 1960s and 70s. Consistent with Porter’s account, these measures emerged in 

applied rather than theoretical branches of social and behavioral science, and they did so not as a 

result of physics envy, but rather as a result of a moral impulse to improve society; 

quantification was intended to make up for perceived deficiencies in unaided human judgment; 

and radical disagreements about the nature of wellbeing did not impede efforts to measure it – 

indeed, in time, there was considerably more agreement about how to measure wellbeing than 

about how to define it. 

 

Keywords: eupathics, happiness, measurement, satisfaction, wellbeing, well-being, positive 

psychology 

 

 

1. Introduction 

It is often suggested that subjective measures of wellbeing – measures designed to represent 

happiness, satisfaction, or other ‚positive‛ or desirable mental states – are a relatively new 

invention. Bruno S. Frey and Alois Stutzer, for example, write: ‚Recently, great progress has 

been achieved in economics: happiness has been seriously measured, and many of its 

determinants have been identified‛ (Frey & Stutzer, 2000, p. 145). Two years later, the same 

authors note that economists traditionally have given little attention to questions of happiness, 

and add: ‚In the past few years the situation has changed: A number of economists see an 

advantage in measuring subjective well-being as expressed by individuals themselves‛ (2002, 

p. vii). Similarly, it is often suggested that the so-called science of happiness – the body of 

empirical generalizations developed using subjective measures – constitutes a new 

development. This suggestion is evident, e.g., in the subtitle of Richard Layard’s book 

Happiness: Lessons from a new science (2005), and it is supported by authors like Martin E. P. 

Seligman, who grumbles: ‚For the last half century psychology has been consumed with a 

single topic only – mental illness‛ (Seligman, 2002, p. ix).  

The impression that the systematic study of happiness and the like constitutes a recent 

development is partially due to its association with the positive psychology movement (cf. 



The evolution of eupathics 

Angner 

 

www.internationaljournalofwellbeing.org 5 

Seligman & Csikszentmihalyi, 2000). This movement, which emerged in the 1990s in large part 

due to Seligman’s efforts, is motivated by the belief that traditional psychology has spent an 

inordinate amount of time examining psychological pathology. According to positive 

psychologists, traditional psychology overlooked ‚positive‛ mental states because they were 

seen as derivative from or less authentic than negative ones, and therefore less worthy of study; 

in contrast, positive psychology proceeds from the assumption that positive mental states are 

no more derivative and no less authentic than negative ones, and consequently worthy of 

attention in their own right (Seligman, 2002; Seligman & Csikszentmihalyi, 2000). Because of 

their implicit or explicit association with positive psychology, which as a self-conscious 

subdiscipline of psychology is a new development, it may seem as though subjective measures 

of wellbeing are new too.  

The goal of this paper is to trace the historical roots of measures of happiness, satisfaction, 

and the like. My thesis is that such measures, far from being a novel invention, have a long and 

rich history, and that this history conforms to the general account of measurement in social and 

behavioral science offered by Theodore M. Porter (1995). The appearance of a science of 

happiness was perhaps marked by a 1917 article promoting a discipline of ‚eupathics,‛ defined 

as the study of ‚the well being of the normal‛ (Myerson, 1917, p. 344). Modern efforts to develop 

measures of happiness, etc., are part of an uninterrupted research stream going back at least to 

the 1920s and 1930s. Subjective measures emerged in marital success studies, educational 

psychology, and personality psychology in the 1920s and 30s, as tools developed for the study 

of personality were applied to happiness and satisfaction; the measures were further shaped by 

the epidemiology of mental health, gerontology, and the social indicator movement in the 1960s 

and 70s, as measures of happiness and satisfaction were recruited as proxies for mental health 

and wellbeing in large, representative samples. Consistent with Porter’s account, subjective 

measures of wellbeing emerged in applied rather than theoretical branches of social science, 

where they were developed not as a result of physics envy but of a moral impulse to improve 

society; quantification was intended to make up for perceived deficiencies in unaided human 

judgment; and radical disagreements about the nature of wellbeing did not impede efforts to 

measure it – indeed, in time, there was considerably more agreement about how to measure 

wellbeing than about how to define it.  

As an exercise in what Hasok Chang (2004) has called history of science as complementary 

science – that is, the ‚continuation of science by other means‛ (p. 249) – the history of subjective 

measures of wellbeing should be of interest not just to historians of psychology, but to 

practicing social and behavioral scientists as well as to anyone interested in the application of 

the science of happiness to public policy. As the quotes in the first paragraph above indicate, 

even highly prominent researchers of happiness and satisfaction appear unaware of the 

longevity and richness of the research stream of which they are part. The story offers further 

evidence for Chang’s thesis that history can serve to recover knowledge – about natural and 

social phenomena as well as about previous efforts to measure them – that appears to have 

been forgotten; to develop critical awareness of contemporary efforts, their prospects and 

limitations; and to suggest novel experimental and theoretical developments.  

 

2. The emergence and evolution of subjective measures  

In this section, I outline how measures of happiness and satisfaction emerged in educational 

psychology, marital success studies, and personality psychology in the 1920s and 30s, and how 

those measures were further shaped by the epidemiology of mental health, gerontology, and 

the social indicator movement in the 1960s and 70s (see Appendix A for a timeline). As my 



The evolution of eupathics 

Angner 

 

www.internationaljournalofwellbeing.org 6 

starting point, I take the references in a prominent contemporary review article (Diener, Suh, 

Lucas, & Smith, 1999) as well as the sources quoted therein; for the early studies, I have relied 

heavily on the review article by Warner Wilson (1967). I group the studies by the discipline in 

which they appeared in order to emphasize the broad movements that generated and sustained 

an interest in happiness, and the like, as well as in the methods required to measure such 

things. This is not to suggest that work in different disciplines was entirely independent: in fact, 

as we will see, there was a fair amount of cross-pollination. Furthermore, I do not mean to 

suggest that the historical sketch below is in any interesting sense complete: given how rich the 

history turns out to be, and given my focus on broad trends, I can offer only brief sketches of 

individual figures, projects, measures, and methods, and hint at the cultural, intellectual, 

academic, social, and economic context of which they were part.  

 

2.1 Marital success studies  

Two of the earliest studies on happiness were penologist and social worker Katharine Bement 

Davis’s Factors in the sex life of twenty-two hundred women (1929) and psychiatrist G. V. 

Hamilton’s A research in marriage (1929). Davis, who received her Ph.D. in economics, was 

interested in gathering ‚adequate data as to both the physical and mental facts of the sex life of 

the normal individual‛ (Davis, 1929, p. ix). She also wanted to explore correlations between 

facts about women’s sex lives and other aspects of their lives. Thus, one of the questions given 

to the roughly 2,200 subjects was the following: ‚Do you consider your life on the whole (a) 

happy, satisfactory, successful; (b) unhappy, unsatisfactory, unsuccessful? In each case why?‛ 

(p. 89). Hamilton’s study was similar in spirit, except that his subjects were married couples. He 

gave a window into what motivated these studies when he wrote:  

My standpoint is that of the psychiatrist who believes that subjective 

phenomena, as these are experienced by the persons who report their 

occurrence, do not need to be translated into anything else in order to be dealt 

with as objectively as we deal with all other biological phenomena (Hamilton, 

1929, p. xi).  

Specifically, Hamilton wanted to know how satisfied his subjects were with their marriages. To 

this end, he used responses to a set of questions about satisfaction to compute a ‚satisfaction-

grade,‛ which he also referred to as an ‚index of spousal satisfaction/dissatisfaction‛ (pp. 78-

79). The satisfaction grade was thought to represent what Hamilton called ‚marital success‛ (p. 

8).  

The work of Davis and Hamilton clearly inspired Lewis M. Terman, Stanford professor of 

education and psychology, proponent of large-scale standardized testing, and famous above all 

for the Stanford-Binet intelligence test (Hilgard, 1957). In a 1938 book called Psychological factors 

in marital happiness, Terman described his project as follows: ‚We have selected as the theme of 

our study that aspect of the successful marriage which may be designated as marital happiness, 

and we wish to ascertain, if possible, what psychological factors are demonstrably associated 

with this state‛ (Terman, 1938, p. 2). To this end, Terman distributed questionnaires to 792 

couples from the ‚middle and upper-middle classes of urban and semiurban Californians‛ (p. 

13). He constructed a ‚marital happiness score‛ on the basis of ‚(1) subjective ratings of the 

happiness of the marriage; and (2) factual information on husband-wife agreement or 

disagreement about various matters,‛ in which the subjective rating ‚was allowed a heavy 

weighting‛ (p. 3). For the subjective rating, respondents were given the question ‚Everything 

considered, how happy has your marriage been?‛ and asked to ‚Draw a circle around 1, 2, 3, 4, 



The evolution of eupathics 

Angner 

 

www.internationaljournalofwellbeing.org 7 

5, 6, or 7,‛ where the options were ‚1 = Extraordinarily happy,‛ ‚2 = Decidedly more happy 

than the average,‛ ‚3 = Somewhat more happy than the average,‛ and so on (p. 440).  

A decade after Davis (1929), but drawing above all on Terman (1938), sociologists Ernest W. 

Burgess and Leonard S. Cottrell published a study called Predicting success or failure in marriage 

(1939). Their study had three goals: to define ‚marriage adjustment,‛ to identify what factors 

were associated with marital success or failure, and to determine whether it was possible to 

predict ahead of time which marriages would lead to happiness and which would not (Burgess 

& Cottrell, 1939, p. 15). Burgess and Cottrell motivated their study of marital success by 

arguing that marital adjustment and incompatibility had become a social problem and 

therefore of public concern (p. 1). Defining their terms, the authors wrote:  

A well-adjusted marriage from the point of view of this study may then be 

defined as a marriage in which the attitudes and acts of each of the partners 

produce an environment which is favorable to the functioning of the personality 

of each, particularly in the sphere of primary relationships (Burgess & Cottrell, 

1939, p. 10).  

In their study, 526 couples were given the following prompt: ‚Appraisal of marriage: very 

unhappy.....; unhappy.....; average.....; happy.....; very happy.....‛ and asked to check the 

relevant box (Burgess & Cottrell, 1939, p. 422). In practice, the authors used happiness as the 

criterion by which they judged both adjustment and success (p. 30).  

In sum, Davis, Hamilton and their followers were interested in a cluster of concepts – 

including ‚happiness,‛ ‚adjustment,‛ ‚success,‛ and ‚satisfaction‛ – which by and large were 

treated as interchangeable, and in the association of those concepts with personality, 

background, and sexual factors. These authors clearly believed that ‚subjective phenomena‛ 

were worthy of study in their own right and that they could be explored directly – without 

being ‚translated into anything else,‛ as Hamilton (1929, p. xi) put it – by means of 

straightforward questions about respondents’ happiness or satisfaction. The description of 

inadequate marital adjustment as a social problem and of public concern underscores the fact 

that they thought the results of their studies could be helpful in alleviating social problems. 

Interestingly, some of these researchers appear to have been more concerned with the 

happiness or satisfaction of the marriage than with that of the people in it (Burgess & Cottrell, 

1939, p. 422; Terman, 1938, p. 440). This literature might have been motivated not primarily by 

a desire to make people happy, but by a desire to ensure a sufficiently high population growth 

rate.  

 

2.2 Educational psychology  

Perhaps the most influential early study of happiness, however, was Goodwin Watson’s 

‚Happiness among adult students of education‛ (1930). Watson, who was a professor of 

education at Columbia University’s Teachers College, introduced his topic in the following 

way:  

No human quest may claim a larger following that that for happiness and 

satisfaction in life. Even the highest ethics tends to justify itself by its 

contribution to human happiness. The Beatitudes themselves constitute an 

attempt to formulate the conditions of the most blessed existence. Certainly any 

educational program in modern times is likely to be justified only in terms of its 

direct or indirect contribution to human happiness. It becomes, therefore, 

extraordinary almost beyond belief that so few attempts have been made to 



The evolution of eupathics 

Angner 

 

www.internationaljournalofwellbeing.org 8 

apply the techniques of psychological study to the understanding of happiness 

(Watson, 1930, p. 79).  

What Watson called ‚the highest ethics‛ might have referred to Aristotle’s notion of eudaimonia 

or to the utilitarians’ maxim of the greatest happiness for the greatest number, and ‚*the+ 

Beatitudes‛ to the blessings said by Jesus in the Sermon on the Mount. Watson clearly 

disapproved of the fact that psychologists had not explored systematically how to achieve 

greater happiness – as instructed by philosophical and religious authorities – and he set out to 

make up for the deficiency.  

Watson distributed anonymous questionnaires to 388 graduate students of education, and 

used a wide variety of measures to elicit degrees of happiness. As we will see in Section 3.4, 

some of the measures were remarkably sophisticated. Examining the responses, Watson noted 

that subjects in general ‚tended to think of themselves as above the average on each happiness 

item‛ (p. 84). Subjects who scored low to average in aggregate happiness ‚appeared to believe 

that they gave an impression of greater happiness than was justified,‛ whereas those who 

scored high ‚felt no sense of masquerade‛ (p. 87). Watson also studied the correlation between 

aggregate happiness and subjects’ personality traits and living conditions. He found that 

happiness was not predicted by IQ scores (p. 88), age, whether the subject came from a small 

town, physical defects, number of siblings, parents’ demand for obedience, having parents 

older than 40, whether the mother had a career, parents’ divorce or separation, school marks, 

extra-curricular activities, crushes, a wise sex education, or (perceived) popularity with the 

opposite sex, among other things (pp. 94-96). Perceived harmony between parents did predict 

happiness, however (p. 94), as did (among other things) a preference to spend a few hours with 

the most delightful known companion of the same sex over one of the opposite sex, an ability to 

give a lecture to high school students about sex, an ability to administer a large group of 

people, an absence of strong fears, health during adolescence, an ability to fuse well, an absence 

of shyness or timidity, and marriage (p. 97).  

Another important early contribution to the literature on happiness was a thesis written by 

Randolph C. Sailer under Watson’s supervision. The thesis was published as Happiness self-

estimates of young men (Sailer, 1931). While Sailer largely followed the approach taken by 

Watson (1930), he was also inspired by Hamilton (1929) and others (Sailer, 1931, p. 7). This fact 

indicates that there was cross-pollination at an early stage: researchers with different 

backgrounds were not ignorant of their predecessors’ efforts and in fact built upon them. 

Distributing questionnaires to 500 young men across the United States, Sailer found among 

other things an association between happiness on the one hand and religion as well as 

sociability on the other (p. 99). He added: ‚An especially notable relationship, the largest found 

in the entire study, connected those reporting themselves ‘more even, on a level’ rather than 

‘more up-and-down, now happy, now depressed’ with the happier groups‛ (pp. 98-99). 

In the mid-1930s, a number of studies followed Watson and Sailer in exploring the 

relationship between various aspects of personality and happiness scores. Many of these 

studies found a relationship that turned out to be surprisingly weak. Thus, Geo rge W. 

Hartmann’s 1934 study ‚Personality traits associated with variations in happiness‛ found little 

correlation between personality traits and experienced happiness (Hartmann, 1934, p. 211). 

Hartmann, who was a psychologist at the Pennsylvania State College, used the term felicity 

instead of happiness (p. 209). Similarly, Percival M. Symonds’s ‚Happiness as related to 

problems and interests‛ (1937) concludes: ‚Happy and unhappy *people+ are remarkably alike 

in their problems and interests. The unhappy do not have peculiar problems but make less 



The evolution of eupathics 

Angner 

 

www.internationaljournalofwellbeing.org 9 

satisfactory adjustments to their problems‛ (Symonds, 1937, p. 293). Like Watson and Sailer, 

Hartmann and Symonds both had ties to the Teachers College at Columbia University.  

In sum, educational psychologists like Watson, Sailer, Hartmann, and Symonds agreed 

with Davis, Hamilton, and their followers that subjective phenomena like happiness and 

satisfaction were worthy of study in their own right. In contrast to some students of marital 

success, Watson et al. were unambiguously interested in happiness, satisfaction, and felicity – 

which they appear to have treated as interchangeable – as a property of the individual person 

rather than his or her marriage. Unlike the authors who preceded them, Watson and his 

followers used a wide variety of sometimes remarkably sophisticated measures to assess the 

degree of happiness exhibited by their subjects. As we will see in Section 3.4, the fact that they 

used a variety of measures allowed them to compare the performance of different measures 

and to argue that simple self-report measures were both reliable and valid and could often 

substitute for more extensive batteries of questions. That line of work set the tone for much of 

the research that followed.  

 

2.3 Personality psychology  

Though efforts to measure happiness and satisfaction have roots in marital success studies and 

educational psychology, the discussion so far also implicates the discipline of personality 

psychology. According to a modern definition: ‚Personality psychology is the scientific study 

of the whole person. The goal of personality inquiry is to provide a scientifically credible 

account of human individuality‛ (McAdams, 2001, p. 11308). At the time Davis, Watson, and 

their followers started studying happiness, personality psychology was in the process of 

emerging as an independent subdiscipline of psychology. While the effort to develop reliable 

methods for studying differences in personality traits across people and over time was not 

completely new, the ‚1921-1938 period was a time of intense research activity,‛ by the end of 

which the field of personality psychology was officially established (Winter & Barenbaum, 

1999, pp. 8-9). The event is marked by the appearance of the first authoritative textbook 

(Allport, 1937) and a landmark study (Murray, 1938) (cf. McAdams, 2001, p. 11309). 

Interestingly, Murray went on to direct the work of Herbert Jeremy Goldings at the Harvard 

Psychological Clinic, to whom we will return in the next section (cf. Goldings, 1954, p. 30). 

From its beginnings in the aftermath of World War I, personality psychology was characterized 

by an emphasis on measurement and psychometrics, and by a desire to be useful to 

corporations and governments (Winter & Barenbaum, 1999, p. 5). Hence:  

Building on the pioneering work of Francis Galton and Alfred Binet on mental 

testing and spurred by the mobilization of large military forces in World War I, 

psychologists began to invent self-report, multi-item tests to assess individual 

differences in personality functioning (McAdams, 2001, p. 11309).  

The hope was that a new exact science of personality would ‚furnish assistance to a corporate 

culture and a government suddenly confronted by dramatic changes and the need to ‘manage’ 

and control an American population that had suddenly become larger, more diverse, and 

‘difficult’‛ (Winter & Barenbaum, 1999, p. 5).  

Scientific and methodological developments relating to happiness and satisfaction during 

the late 1920s and 1930s are usefully seen against this background. The happiness research 

surveyed above was obviously driven by a desire to measure and account for personality 

characteristics across individuals. It was also apparently motivated by an ambition to enhance 

the happiness of people, or their marriages, and to help governments alleviate social problems. 



The evolution of eupathics 

Angner 

 

www.internationaljournalofwellbeing.org 10 

Seeing the measurement of happiness as an outgrowth of personality psychology is useful for 

answering Watson’s question at the very beginning of his 1930 paper (cited in Section 2.2). 

Watson asked why it had taken so long for psychologists to explore empirically how happiness 

is distributed, what factors predict it, and how one should go about promoting it. The link with 

personality psychology goes a long way toward answering this question: it was not unti l after 

Galton, Binet, and the experiences of World War I that psychologists developed the confidence 

that they could reliably measure personality characteristics like happiness.  

In conclusion, the emergence of personality psychology as an independent subdiscipline of 

psychology was a critical step in the history of measures of happiness and satisfaction and the 

science of happiness. It was only with the establishment of personality psychology, thanks to 

people like Allport and Murray, that many psychologists developed the confidence that they 

could reliably and validly measure attributes of the person and that the results could be useful 

to governments and corporations. This hypothesis explains why, before the 1920s, few 

psychologists attempted to systematically measure happiness and satisfaction: the 

psychologists of the time had not yet convinced themselves that it was possible to measure 

attributes of the person. The hypothesis also explains why during the 1930s many psychologists 

thought that such a study was not only possible but long overdue. Given that for philosophical, 

religious, or other reasons they thought that happiness was important to understand and 

promote, the conviction that it was possible to measure happiness quickly produced a desire to 

measure it.  

 

2.4 Synthesis and further development  

After these developments, in short order, there emerged a number of studies that drew equally 

on the results from marital happiness studies and educational psychology. For example, 

Hornell Hart’s inventive contribution Chart for happiness (1940) explicitly drew on both.1 Hart, a 

professor of sociology at Duke University, asked: ‚One of the basic purposes of mankind is to 

be happy. Can recent advances in scientific thinking tell us more and more effectively what to 

do in order to be happy, and in order to help make our fellow human beings happy?‛ (p. 16). 

The book was written both for those who ‚sense the fact that they are not living on as full a tide 

of happiness as they might attain‛ and for ‚psychologists, sociologists, psychiatrists, 

physicians, educators, social workers, pastors‛ and other professionals who are concerned with 

the happiness of others (pp. v-vi). Hart paid more attention to the nature of happiness than his 

predecessors, offering not one but two ‚operational‛ definitions of the term: according to the 

one, people are happy insofar as they are prone to saying sincerely that they are happy; 

according to the other, people are happy insofar as they are in whatever state of consciousness 

that they seek to attain or maintain (pp. 182-184; cf. Section 3.3 below).  

Hart was deeply impressed by measurement instruments used in other scientific disciplines 

and hoped to develop something equally useful for happiness. Having discussed intelligence 

tests, the fever thermometer, and the electrocardiograph, he went on to write:  

Chart for Happiness presents and explains another such invention, the 

Euphorimeter, produced by the researches of various social scientists in various 

universities and colleges. This new instrument is to measure, not intelligence, or 

fever, or cardiac pulsations, but the happiness of those who submit themselves 

                                                
1 Specifically, Hart discussed data from Hamilton (1929), Watson (1930), Sailer (1931), Terman (1938), and Burgess 

and Cottrell (1939) (cf. Hart, 1940, pp. 173-174). Hart was well versed in the research that preceded him.  



The evolution of eupathics 

Angner 

 

www.internationaljournalofwellbeing.org 11 

to it. It is to diagnose, not mental aptitudes, nor physical disease, but the causes 

of human anguish (p. v).  

Hart wanted happiness scores to be measurable on an ‚objective scale < which will have 

meaning for ordinary people‛ (p. 177). Hence, he normalized the scores and assigned a score of 

zero to the point that divides happiness from unhappiness and 100 to the average happiness of 

all persons tested (pp. 175-176). As Hart described it:  

The scale of this instrument is somewhat like the scale of a centigrade 

thermometer. The zero on the Euphorimeter is the dividing point between 

happiness and unhappiness. From zero down, the scores mean deeper and 

deeper unhappiness. From zero up, the scores indicate greater and greater 

happiness. The 100-degree point is set by the average happiness of the general 

population (p. 19).  

The unit of the Euphorimeter is called a Euphor-unit, and is defined as ‚one one-hundredth of 

the difference between the zero point and the average happiness score‛ (p. 178). Hart’s method 

for normalizing happiness scores allowed him to compare the results of various studies, as 

shown in Figure 1. 
 

 

Figure 1. The Euphorimeter scale, with some typical scores.2 

 

For purposes of his own research, Hart developed two separate Euphorimeters. Because 

people’s happiness fluctuates from moment to moment, Hart wrote, ‚it has been necessary to 

devise two types of Euphorimeters – one to measure happiness ‘at the moment,’ the other to 

measure as closely as possible the general level of happiness or unhappiness on which one lives 

in the long run‛ (p. 21). One advantage of the At-the-Moment Euphorimeter was that it allowed 

him to develop a dense record of participants’ happiness levels over time. Hart used this 

Euphorimeter to draw charts showing how the happiness of three subjects, one who was falling 

in love, one whose mother was dying, and one who wondered why not everybody committed 

suicide, evolved over time (see Figures 2, 3, & 4 below). 

                                                
2
 Figures 1, 2, 3, & 4 are reproductions based on Hart (1940, pp. 20-24). 

500 

100 

0 

-500 

Young man who was falling in love 

Recently married couples 

Average of 2,200 people tested WPA workers 

Couples who have contemplated divorce 

Prisoners working on roads 

Young man whose mother was dying 



The evolution of eupathics 

Angner 

 

www.internationaljournalofwellbeing.org 12 

 

 

Figure 2. Happiness fluctuations of a student who was falling in love. 

 

 
Figure 3. Happiness fluctuations of a student whose mother was dying. 

 Preached successful sermon; met enjoyable girl  

 
Picnic – wonderful fellowship; all caught up on studies  

700 

600 

500 

400 

300 

200 

100 

0 

-100 

-200 

-300 

-400 

-500 

-600 

-700 
19  21  23  25  27  29  31  2  4  6  8  10  12   

  20  22   24  26  28   30  1  3  5  7  9   11  13 

   M arc h   A p r il 

Mother died; he felt relieved 

that she was out of suffering  

Called home by message  

that mother cannot live  

School work  

Mother much worse  

Not sure he can ‚date‛ her  

Gets very significant letter from her  

Gets very significant letter from her  

Unexpectedly  

fails to get ‚date‛  

Gets first ‚date‛  

Greatly attracted   

  by girl  

700 
 

 
600 

 

 
500 

 

 
400 

 

 
300 

 

 
200 

 

 
100 

 

 
0 

6  8  10  12  14  16  18  20  22  24  26  28  30  1  3  

  7  9   11  13  15  17  19  21  23  25  27  29  31  2  4  

       M arc h        A pr il 



The evolution of eupathics 

Angner 

 

www.internationaljournalofwellbeing.org 13 

 
Figure 4. Happiness fluctuations of a college boy who wondered why everybody did not 

commit suicide. 

 

Hart found the case of the suicidal college student particularly interesting. Taking the test 

apparently made the student realize that he was unusually unhappy and for no good reason; 

by tracking his Euphorimeter readings carefully, it took him no more than a month to rise from 

the depths of despair to above-average happiness. Hart concluded: ‚Perhaps you can make 

equally great progress, if you also are unhappy‛ (p. 25). 

In 1954, under Murray’s supervision, Herbert Jeremy Goldings of the Harvard 

Psychological Clinic published another study, which drew on research both on marital 

happiness and educational psychology. Goldings suggested that he wanted to tap into ‚a 

commercially vigorous popular concern with happiness (books, magazines, etc.) indicating, 

perhaps, that in American culture people are not happy, or not as happy as they want to be, or 

not as happy as they feel they should be‛ (Goldings, 1954, p. 30). Goldings’s article is 

interesting in part because it contains an explicit discussion of ‚the intrinsic nature and 

characteristics‛ of happiness (p. 31). He defined happiness as a physiological state, though one 

that has consequences that are accessible through introspection. His article is also interesting 

because it uses new methods to measure happiness. Other than using measures adapted from 

Watson (1930) and Sailer (1931), Goldings asked his subjects to rate the happiness and 

satisfaction of 30 photographs of faces with ambiguous facial expressions, and proposed that 

300 
 

 
200 

 

 
100 

 

 
0 

 

 
-100 

 

 
-200 

 

 
-300 

 

 
-400 

 

 
-500 

 

 
-600 

9          9   11  13  15  17  19  21  23  25   

                                   12  14  16  18  20  22  24 

Feb r uar y       M arc h 



The evolution of eupathics 

Angner 

 

www.internationaljournalofwellbeing.org 14 

the result could serve as an ‚indirect‛ measure of happiness, on the theory that subjects would 

project their own happiness onto the ambiguous facial expressions (Goldings, 1954, p. 33; cf. 

Section 3.4 below). Using 20 Harvard undergraduates as subjects, Goldings confirmed 

Watson’s finding that subjects tend to rate their own happiness as greater than the average 

(Goldings, 1954, p. 36). Goldings also found ‚a highly significant degree of agreement among 

experienced judges’ ratings of happiness‛ and a high correlation between avowed happiness 

and judges’ ratings (p. 46). However, because it did not behave as anticipated, he ended up 

rejecting the indirect measure as a measure of happiness.  

In brief, Hart and Goldings were well aware of their predecessors in marital success studies 

and educational psychology, and they sought ways to synthesize and build upon that research. 

Both believed that individual happiness was worthy of study in its own right, that the scientific 

approach to happiness had particular power and potential, that self-report measures of 

happiness were adequate for a wide variety of purposes, and that scientific studies could help 

people become happier. By comparison with their predecessors, both Hart and Goldings 

offered a more explicit discussion about the nature of happiness. They also continued the 

educational psychologists’ efforts to develop and test novel ways to assess happiness. Thus, 

Hart developed a procedure, inspired by the centigrade thermometer, for standardizing 

happiness scores. Hart also started tracking people’s happiness over time, which is particularly 

interesting in light of the affinities with the much more recent work by Daniel Kahneman and 

co-authors (e.g., Redelmeier & Kahneman, 1996). Meanwhile, Goldings developed a new 

projection-based indirect measure. Not all their efforts succeeded: because Goldings’s indirect 

measure failed to behave as expected, he abandoned the project of using such measures as 

measures of happiness. Both Hart and Goldings expressed a desire to reach a wider audience, 

and perhaps to tap into what Goldings (1954, p. 30) had called the ‚commercially vigorous 

popular concern with happiness.‛  

 

2.5 The epidemiology of mental health  

In the late 1950s and early 60s a number of authors with a background in the epidemiology of 

mental health recruited measures of happiness and satisfaction for use as measures of mental 

health in large, sometimes representative samples. This move, as we will see shortly, appears to 

have had an enormous impact on the use and interpretation of measures of happiness and 

satisfaction.  

A landmark study in the epidemiology of mental health is Americans view their mental health, 

by Gerald Gurin, Joseph Veroff, and Sheila Feld (1960). As the title indicates, Gurin and his co -

authors were primarily interested in how individuals themselves – as opposed to mental health 

experts – saw their mental health (Gurin et al., 1960, pp. 3-4). The authors adopted a multiple-

criterion perspective: they believed that mental health was best measured by using a battery of 

criteria, one of which was whether people felt happy. Thus, Gurin et al. did not identify 

happiness and mental health, but saw the former as an important component or indicator of the 

latter. One of the questions they asked was the following: ‚Taking things all together, how 

would you say things are these days – would you say you’re very happy, pretty happy, or not too 

happy these days?‛ (p. 411). Unfortunately, although they discussed what the subjects thought 

contributed to their happiness, Gurin et al. said little about what they meant by the term (pp. 

22-28). Because they wanted a representative sample, Gurin et al. enrolled 2,460 randomly 

selected, non-institutionalized American adults (p. 3). In this sample, Gurin et al. found an 

association between unhappiness and pessimism and uncertainty about the future (p. 35), but 

no connection between the sex of the subject and his or her happiness (p. 42). They found that 



The evolution of eupathics 

Angner 

 

www.internationaljournalofwellbeing.org 15 

older people were less happy than younger (p. 44), and that those with more education were 

happier than those with less (p. 46). They also found that happiness was positively related to 

income (p. 216).  

Another important contribution to the epidemiology of mental health is Norman M. 

Bradburn and David Caplovitz’s book Reports on happiness: A pilot study of behavior related to 

mental health (1965), which built on Gurin et al. (1960). According to Bradburn and Caplovitz, 

this study was ‚an effort to develop, for psychological and behavioral phenomena, time-series 

studies comparable to those that are commonplace in economics and demography‛ (Bradburn 

& Caplovitz, 1965, p. 1). As the following quote indicates, they were particularly interested in 

‚psychological well-being‛:  

Its long-range objective is to conduct periodic inventories of the psychological 

well-being of the nation’s population. From such inventories it is possible to 

determine for the first time the extent to which feeling states of the population 

are affected by major social trends, national and local crises, and changes in the 

economic and social structures, as well as by patterned events in the life cycles of 

individuals (p. 1).  

It appears that the authors identified ‚psychological well-being‛, ‚mental health‛, and 

‚happiness‛ into a one-dimensional entity. In Bradburn and Caplovitz’s words:  

The underlying assumption of this research is that there is a dimension variously 

called mental health, subjective adjustment, happiness, or psychological well-

being, and that individuals can be meaningfully described as being relatively 

high or low on such a dimension (p. 1). 

The authors were comfortable using self-report measures, as they wrote: ‚The most fruitful 

starting point in developing an instrument to measure a dimension of mental health seemed to 

be people’s own estimates of their level of psychological well-being or distress‛ (p. 5). Using 

the specific measure employed by Gurin et al. (1960), Bradburn and Caplovitz administered 

questionnaires to 2,006 members of 450 households in four rural Illinois communities 

(Bradburn & Caplovitz, 1965, pp. 3-5) and concluded among other things that happiness was 

positively correlated with education and income, negatively correlated with age, and 

uncorrelated with sex (p. 56).  

In sum, mental health professionals like Gurin, Bradburn and Caplovitz were concerned 

with the causes, distribution, and promotion of mental health in the population, and decided 

that measures of happiness and satisfaction could be recruited for the purpose. They thought it 

legitimate to use a measure of happiness as a measure of mental health because they 

considered happiness either identical to, a component of, or an indicator of mental health or 

psychological wellbeing. To assess individual happiness, these authors were comfortable using 

a single straightforward question, answered on a three-point scale, which remains in use today. 

As compared to previous research, the effort of the mental health professionals was different in 

important respects. Above all, Gurin et al. wanted to know the distribution of happiness in the 

population as a whole, so they started using large, random, and therefore representative, 

samples from the national population; Bradburn and Caplovitz wanted to know not only how 

happiness was distributed in the population, but also how the distribution developed over 

time, so they started collecting time series data, much like Hart (1940) had. This approach 

would be picked up and further explored by representatives of the social indicators movement, 

to which we will return in Section 2.8. In recent years, of course, the effort to track the 

happiness and satisfaction of the nation over time has received a great deal of attention under 



The evolution of eupathics 

Angner 

 

www.internationaljournalofwellbeing.org 16 

headings like ‚national indicators of subjective well-being‛ and ‚national well-being accounts‛ 

(e.g., Diener, 2006; Kahneman, Krueger, Schkade, Schwarz, & Stone, 2004).  

 

2.6 Review and call for more research  

In 1967, University of Alabama psychologist Warner Wilson published the first major review 

article of research on happiness, or as he said (borrowing a term of Goldings’s), ‚avowed 

happiness.‛ Wilson reviewed the research discussed above, his 1960 dissertation in the area, 

and more. On the basis of this meta-analysis, Wilson concluded that ‚avowed happiness can be 

determined reliably,‛ and that the ‚facts < support the validity of self-ratings‛ (1967, pp. 294-

295). On the basis of the studies he reviewed, Wilson inferred: ‚The happy person emerges as a 

young, healthy, well-educated, well-paid, extroverted, optimistic, worry-free, religious, 

married person with high self-esteem, high job morale, modest aspirations, of either sex and of 

a wide range of intelligence‛ (p. 294). Under ‚Suggestions for further research,‛ Wilson offered 

the following: ‚Comparability of results might be increased if all investigators used the same 

measure. The designations ‘very happy,’ ‘pretty happy,’ and ‘not too happy’ used by Gurin et 

al. (1960) and by Bradburn and Caplovitz (1965) seem as good as any‛ (Wilson, 1967, pp. 304-

305). 

Wilson ended his article by calling for more experimental research, in which the happiness 

of subjects is manipulated through therapy or other means. ‚Studies involving direct attempts 

to manipulate the well-being of individuals are most desirable,‛ he maintained (p. 305). He 

continued:  

A story is told of how a small group of college men increased the poise and 

popularity, and presumably the happiness, of a female student by going out of 

their way to respond to her as though she were attractive. The possibility of 

systematically developing and applying such principles and techniques seems 

exciting indeed! (p. 305) 

While nobody, to my knowledge, has performed this particular experiment under controlled 

conditions – presumably because of ethical constraints – more recent happiness scholars have 

started performing controlled studies examining the causes of happiness. Many of them are 

interested in developing interventions that can reliably be used to enhance people’s happiness 

(cf. Lyubomirsky, 2008).  

 

2.7 Gerontology  

Around the same time, a number of studies on happiness and satisfaction had appeared in the 

field of gerontology (Larson, 1978). According to David L. Adams (1971), a concern with 

‚individual well-being‛ was one of the factors contributing to the development of gerontology 

in the first place (p. 64). He continued: ‚However, the difficulty of trying to assess ‘individual 

well-being’ has resulted in a variety of concepts, definitions, and measurements‛ (p. 64). The 

concepts included ‚morale,‛ ‚successful aging,‛ and ‚personal adjustment.‛ From the start, 

according to Bernice L. Neugarten, Robert J. Havighurst, and Sheldon S. Tobin (1961), there 

were two different approaches to measuring psychological wellbeing or wellbeing in 

gerontology. The one, according to Neugarten et al., focused upon ‚the overt behavior of the 

individual and utilizes social criteria of success or competence‛ (p. 134). This approach was 

criticized, however, for relying too heavily on the assumption that social competence and 

participation were closely associated with wellbeing, and for imposing on the subjects the value 

judgments of the investigators (p. 134). The other approach, which emerged at least in part as a 



The evolution of eupathics 

Angner 

 

www.internationaljournalofwellbeing.org 17 

reaction to the first, relied instead on an ‚internal frame of reference‛ in which attributes like 

success and competence figured only indirectly. Hence: ‚Here the variables to be measured 

have been the individual’s own evaluations of his present or past life, his satisfaction, or his 

happiness‛ (p. 134). In this approach:  

The assumptions are, whether or not explicitly stated, that the individual himself 

is the only proper judge of his well-being; that the value judgments of the 

investigator can thus be minimized; and, perhaps most important, that it is not 

appropriate to measure well-being in old age by the same standards that apply 

to middle age, namely, standards based upon activity or social involvement (p. 

134).  

By relying on the individual’s own evaluations, happiness, and satisfaction, in this view, 

researchers could avoid multiple problems associated with assessments of success or 

competence.  

There were several studies that relied on this ‚internal frame of reference‛ (e.g., Lebo, 

1953). Many of the early studies combined elements of the two approaches, and asked 

questions about ‚happiness‛ and ‚feelings of well-being or satisfaction‛ along with questions 

designed to assess degrees of social competence (Pollak, 1948, pp. 66-67). Thus, in an influential 

1949 book, Ruth Shonle Cavan, Burgess, Robert J. Havighurst, and Herbert Goldhamer aspired 

‚to define and to analyse the nature, patterns, and problems of personal adjustment to ageing‛ 

(Cavan, Burgess, Havighurst, & Goldhamer, 1949, p. v). The second author of this book was 

also the first author of a major contribution to the literature on marital happiness (Burgess & 

Cottrell, 1939), which underscores the fact that the different traditions explored here were not 

independent of one another. Cavan et al. asked their 3,000 participants questions about both 

whether they felt satisfied, and whether, for example, they participated in clubs and 

organizations (Cavan et al., 1949, Appendix A).  

The most common focus in gerontology, however, was on ‚life satisfaction‛ (cf. Adams, 

1971; Neugarten et al., 1961). Thus, Arnold M. Rose (1955) studied the life satisfaction of 

married adults whose children had just left the home. Rose asked his subjects: ‚In general, how 

satisfied are you with your life?‛ and offered them a list of five alternative answers: ‚Very 

dissatisfied,‛ ‚Somewhat dissatisfied,‛ ‚About average,‛ and so on (Rose, 1955, p. 15). Rose 

found that women (but not men) who married before the age of 20 or after the age of 30 tended 

to be less satisfied with their lives, that dissatisfied women were less likely to have a paying job 

and more likely to think they spent an inordinate amount of time doing housework, and that 

satisfaction was associated with social participation in both men and women (Rose, 1955, pp. 

18-19). A more explicit treatment of measures of life satisfaction appeared in Neugarten et al. 

(1961). They took satisfaction to be closely related to ‚Zest (vs. apathy); Resolution and 

fortitude; Congruence between desired and achieved goals; Positive self concept; and Mood 

tone‛ (p. 137), but did not specify if these factors were supposed to be components, causes, or 

mere correlates of satisfaction.  

In sum, gerontologists have been interested in wellbeing since the establishment of the 

discipline. While some authors in this tradition spoke about ‚psychological well-being‛ 

(Neugarten et al., 1961, p. 134), they frequently talked about ‚well-being‛ without the modifier 

(cf. Adams, 1971, p. 64; Neugarten et al., 1961, p. 134). Numerous gerontologists appear to have 

turned to happiness and satisfaction as proxies for wellbeing because they thought that by 

doing so they could avoid imposing their own value judgments and standards of wellbeing on 

their subjects; because they wanted to rely, to the greatest extent possible, on the subjects’ own 

standards of wellbeing; and because they felt that the subjects themselves were best situated to 



The evolution of eupathics 

Angner 

 

www.internationaljournalofwellbeing.org 18 

determine to what extent they satisfied those standards. The tension between the two 

approaches to wellbeing measurement in gerontology is interesting because it mirrors the 

contemporary tension between subjective measures of wellbeing and capability-based 

measures. Capability-based measures are designed to reflect not mental states like happiness 

but capabilities and functionings, that is, what a person can be and do in leading a life (Sen, 

2008, p. 24).  

 

2.8 The social indicators movement  

A great deal of interest in subjective wellbeing during the 1960s and 70s grew out of the social 

indicators movement. This movement emerged in the late 1960s as a reaction against the 

widespread use of economic indicators like GDP as measures of ‚the goodness of life‛ 

(Andrews, 1989, p. 401; Campbell, 1976, p. 118). For instance, Angus Campbell (1976) admitted 

that economic measures were ‚easy to count‛ and he suggested that this goes a long way 

toward accounting for their popularity as measures of wellbeing. Yet, he added: ‚None of us 

doubts that economic data have admirable qualities; the question is, How well do they 

represent the quality of national life? How valid are they as measures of the goodness of life in 

this country?‛ (p. 117). Campbell was obviously not enthusiastic about economic measures. In 

contrast, the social indicator movement proposed a set of indicators that, although perhaps not 

as ‚easy to count,‛ would nevertheless be better measures of quality of life. As Frank M. 

Andrews put it:  

The idea called for two key changes from earlier practices. One was an 

expansion in the range of phenomena monitored beyond the traditional 

economic indicators, and an explicit recognition that ‚life quality,‛ however it 

might be designed, involved more than just economic considerations. The 

second change involved an attempt to focus directly on ‚output‛ indicators – 

i.e., indicators that show how well off people actually are – in addition to the 

more traditional ‚input‛ indicators that reflect budget allocations, procedures 

and processes that are presumed to enhance well-being (Andrews, 1989, p. 401).  

The social indicator movement instead generated a wealth of so-called ‚objective indicators,‛ 

that is, statistics concerning life expectancy, quality of food and water, access to medical care, 

level of education, quality of housing, and so on (Campbell, 1976, p. 118). The underlying 

assumption, of course, was that more education, housing, and medical care tended to increase 

the quality of life. As Campbell put it: ‚It is reasonably argued that as the level of education 

rises, the adequacy of medical care improves, the amount of substandard housing is reduced, 

and the purity of the air and water is increased, the quality of life is therewith enhanced‛ (p. 

118).  

Nevertheless, a number of researchers came to think that objective indicators did not go far 

enough. As Campbell wrote:  

[If] we believe, as I assume that most psychologists do, that the quality of life lies 

in the experience of life, then these are surrogate indicators. They describe the 

conditions of life that might be assumed to influence life experience, but they do 

not assess that experience directly (1976, p. 118).  

The problem was that indicators concerning education, housing and health care remained 

‚input‛ indicators, and as such were subject to the same criticism as economic welfare 

measures. Thus, the very argument that led to the rejection of economic indicators appears to 

have led to the rejection of objective indicators as well. In Campbell’s words: 



The evolution of eupathics 

Angner 

 

www.internationaljournalofwellbeing.org 19 

If we are primarily concerned with describing the quality of life experience of 

the population, we will need measures different from those that are used to 

describe the objective circumstances in which people live. We will have to 

develop measures that go directly to the experience itself. These subjective 

measures will surely not have the precision of indicators that are expressed in 

numbers of dollars, units of time, or numbers of square feet, but they will have 

the great advantage of dealing directly with what it is we want to know, the 

individual’s sense of well-being (p. 118).  

Hence, it appears that the logic of their position forced members of the social indicator 

movement to reject an exclusive reliance not just on economic measures but on the wider set of 

‚objective‛ measures as well, and to favor what they called more ‚direct‛ measures of quality 

of life.  

One hugely influential study in this vein is Princeton psychologist Hadley Cantril’s book 

The pattern of human concerns (1965). Cantril was particularly interested in people’s aspirations 

and in the manner in which aspiration levels affected their happiness and satisfaction. His goal 

was to find a method to estimate the aspirations and satisfactions by reference to the subjects’ 

own, possibly changing, standards (Cantril, 1965, p. 21). For this purpose, he invented what he 

called the ‚Self-Anchoring Striving Scale‛ (p. 22). When using the scale, the subject is asked to 

judge how good his or her life is on a graphical, ladder-like device, where the top and bottom 

rungs represent respectively the best and worst possible life he or she could live (p. 23). Cantril 

and his collaborators administered the test to a total of 23,875 subjects in 13 countries. Under 

the heading ‚Who Are the Satisfied?,‛ Cantril discussed what characterized individuals who 

rated themselves high on the Self-Anchoring Striving Scale. Among other things, Cantril found 

that education, income, and occupation were strongly correlated with ladder ratings, that 

‚people living in urban centers‛ rated themselves somewhat higher than ‚those living on the 

land,‛ but that men and women rated themselves about the same (p. 258). 

Campbell, Philip E. Converse, and Willard L. Rodgers opened up their discussion in The 

quality of American life (1976) by pointing to deficiencies both with traditional economic 

measures of wellbeing and with those proposed by the social indicator movement. Their book 

had two objectives. First, the authors wanted to establish ‚a system of reporting of 

psychological data which parallels our current repertoire of social indicators‛ (Campbell et al., 

1976, p. 4). Second, they wished to provide ‚a fuller and truer representation of the state of 

society to those people who are responsible for social decisions‛ (p. 5). Campbell et al. (p. 5) 

drew primarily on the three following sources: Gurin et al. (1960), which they call the ‚first 

major study of the quality of life experience‛; Bradburn and Caplovitz (1965); and Cantril 

(1965). On the basis of personal interviews with a random sample of 2,164 Americans 

(Campbell et al., 1976, p. 511), the authors reported: ‚Persons who are currently single 

generally report a good deal less satisfaction with life than do married persons, and the lack of 

satisfaction shown by women and men who are divorced or separated is quite remarkable 

indeed‛ (p. 36). The young tended to report higher happiness than the old, but the old 

appeared to be more satisfied with life than the young (p. 36). The unemployed and the 

divorced scored ‚conspicuously low‛ (p. 51) and there was a ‚rather strong and regular 

relationship between income and a sense of well-being‛ (Campbell et al., 1976, pp. 51-55).  

Frank M. Andrews and Stephen B. Withey’s book Social indicators of well-being: Americans’ 

perceptions of life quality (1976) was another major contribution. Andrews and Withey placed 

their research firmly in the social indicator movement. They wrote that ‚social indicators of the 

various conditions of human beings, and of the changes that characterize their lives, offer much 



The evolution of eupathics 

Angner 

 

www.internationaljournalofwellbeing.org 20 

that is attractive,‛ but added that ‚in these days of growing interdependence and social 

complexity we need more adequate cues and indicators of the nature, meaning, pace, and 

course of social change‛ (p. 1). Though the two said they chose to concentrate on ‚perceptions 

of well-being,‛ they immediately went on to say: ‚Measurements of individual well-being seem 

to us, and to many others also, a particularly promising place to begin‛ (pp. 6-7). This suggests 

that they in fact did not draw a sharp distinction between wellbeing on the one hand and 

perceptions of wellbeing on the other. The study drew primarily on ‚four nationally 

representative probability samples of the American adult population,‛ for a total of 

approximately 5,500 individuals (p. 24).  

The social indicators movement, then, played a critical role in the development of subjective 

measures of wellbeing. Though prominent people like Andrews and Campbell had initially 

advocated the use of objective measures, the logic of their position pushed them toward more 

‚direct‛ measures of happiness and satisfaction. The argument that subjective measures are 

superior because they are more direct has since become part of a standard defense of subjective 

measures (cf. Angner, 2011). Like at least some of the gerontologists, these researchers thought 

of measures of happiness and satisfaction as measures of wellbeing, without any modifiers. 

This is clear from their use of the term ‚subjective measures of well-being,‛ e.g. in the title of 

Campbell (1976), and from the fact that subjective measures were compared with, and found to 

be superior to, the more established economic welfare measures and the so-called objective 

indicators. These authors continued the development of subjective measures, and some of them 

– like Cantril’s Self-Anchoring Striving Scale – remain in use today. Like the epidemiologists 

we encountered in Section 2.5, authors like Andrews and Campbell aspired to track the 

wellbeing of large, representative samples over time. Unlike its predecessors, the social 

indicator movement unashamedly used subjective measures of wellbeing as macro-level 

indicators of quality of life – that is, as measures of social wellbeing – comparable to the 

standard economic indicators. Since the subjective measures were thought to be superior to 

economic or other objective measures, subjective measures of wellbeing emerged as serious 

substitutes for, e.g., measures of GDP as welfare indicators. This idea remains part of the effort 

to develop ‚national well-being accounts,‛ as described in Section 2.5. 

 

2.9 Summary and discussion 

In this section, I have sketched the emergence and development of measures of happiness and 

satisfaction. Far from being a novel invention, these measures have an uninterrupted history 

that can be traced back at least to the 1920s and 30s. It was in the context of marital success 

studies that researchers like Davis, Terman, and others started systematically studying 

happiness and satisfaction, although some of them were interested in the happiness of the 

marriage, rather than that of the individuals in it. Educational psychologists like Watson were 

unambiguously interested in the happiness and satisfaction of individuals, as opposed to 

marriages; their use of a wide range of remarkably sophisticated happiness measures allowed 

them to argue on empirical grounds that simple self-report measures were both reliable and 

valid and could substitute for more extensive batteries of questions. The confidence that 

attributes of the person could be reliably and validly measured, and that the results could be 

useful to governments and corporations, was due to the emergence of personality psychology 

shortly after World War I. Epidemiologists of mental health started using measures of 

happiness and satisfaction as measures of mental health in large, nationally representative 

samples. Meanwhile, gerontologists used, above all, measures of satisfaction as measures not 

just of mental health, or psychological wellbeing, but of wellbeing without the modifier. The 



The evolution of eupathics 

Angner 

 

www.internationaljournalofwellbeing.org 21 

social indicator movement entrenched the idea that measures of happiness and satisfaction 

could serve as macro-level indicators of quality of life – that is, as measures of social wellbeing 

– and as serious substitutes for, e.g., measures of GDP as welfare indicators.  

It might be objected that the foregoing discussion ignores interesting and important 

contributions to the literature. Indeed, several early contributions that were in important 

respects similar to those discussed above, like Adolf Wohlgemuth (1919), J. C. Flügel (1925), 

and Arthur E. Morgan (1934), have not been included. Obviously, the story told here is 

incomplete: given how long and rich the history turned out to be, some omission is inevitable. 

But it is also worth remembering that my aim was not to identify researchers who 

‚anticipated‛ modern research efforts, but to try to identify the work that was the most 

influential. Discussion of the economists who brought happiness research to the attention of 

mainstream economics, above all Bernard M. S. van Praag (1968) and Richard A. Easterlin 

(1974) has also been omitted. These more recent developments are not discussed because their 

influence begins as this story ends.  

 

3. Porter on measurement in social and behavioral science  

In this section, I will argue that the history of subjective measures of wellbeing from the 1920s 

or 30s, in all essentials, confirms Porter’s general account of the history of measurement in 

social and behavioral science. Porter’s account is most clearly articulated in the book Trust in 

numbers: The pursuit of objectivity in science and public life (1995) but has been discussed and 

developed in a number of forums, including the collection The age of economic measurement 

(Klein & Morgan, 2001) and the special issue of Studies in History and Philosophy of Science (Vol. 

32, No. 4, 2001), edited by E. P. Hamm and Alan W. Richardson. Consistent with Porter’s 

account, I will argue, subjective measures of wellbeing emerged in applied, rather than 

theoretical, branches of social and behavioral science, where they were developed not as a 

result of physics envy but of a moral impulse to improve society; quantification was intended 

to make up for perceived deficiencies in unaided human judgment; and radical disagreements 

about the nature of wellbeing did not impede efforts to measure it – indeed, in time, there was 

considerably more agreement about how to measure wellbeing than about how to define it.  

 

3.1 Subjective measures appeared in applied branches of psychology  

One of Porter’s main theses is that the drive toward quantification was strongest in applied 

rather than theoretical branches of social science, and that it was the result not of physics envy, 

but rather of a moral impulse to understand and improve a changing society (Bateman, 2001, p. 

57; Levy, 2001, p. 724; Porter, 1995, p. viii). The history of the measurement of happiness and 

satisfaction confirms Porter’s thesis: measures of happiness and satisfaction emerged in applied 

branches of psychology, sociology, and medicine, and they appeared not as the result of an 

effort to replicate the successes of quantification in the natural sciences, but rather of an 

ambition to understand the causal antecedents of happiness and satisfaction in order to build a 

happier and therefore better society.  

As we have seen, subjective measures of wellbeing did in fact emerge in applied branches 

of psychology, sociology, and medicine: in marital success studies (Section 2.1), educational 

psychology (Section 2.2) and personality psychology (Section 2.3), as well as in the 

epidemiology of mental health (Section 2.5), gerontology (Section 2.7), and social indicators 

research (Section 2.8). It is true that happiness, satisfaction, and other positive or desirable 

emotional states have begun to attract attention from more theoretical branches of psychology, 

economics, and neuroscience. Hence, a collection like Well-being: The foundations of hedonic 



The evolution of eupathics 

Angner 

 

www.internationaljournalofwellbeing.org 22 

psychology (Kahneman, Diener, & Schwarz, 1999) contains chapters, e.g., on the neuroscientific 

basis for utility appraisals. Yet, this is a relatively recent phenomenon, which does not change 

the fact that the measures originally emerged in applied branches of social and behavioral 

science.  

There is little evidence that proponents of subjective measures suffered from ‚physics 

envy,‛ in the sense of an urge to replicate the successes of quantification in physics and other 

natural sciences. Some of Hart’s examples of successful measurement – including the 

electrocardiograph and the thermometer – did arguably come from the natural sciences; yet, his 

focus was unambiguously on practical applications in the interest of investigating and solving 

social problems. Here is how he introduced his discussion about the power of quantification:  

When intelligence tests were invented a new instrument was placed in the hands 

of workers with human beings – a powerful instrument for diagnosing 

maladjustments, assaying human capabilities, placing students and workers in 

the surroundings where they would be most useful, and investigating various 

social problems. When the fever thermometer was invented a new instrument 

was placed in the hands of physicians – a powerful instrument for diagnosing 

bodily ailments. When the electrocardiograph was invented a new instrument 

for tracing the pulsations of the heart was made available for diagnosis and for 

pointing the way toward remedial treatment (Hart, 1940, p. v). 

Hart also pointed to the contribution of ‚aerial dynamics‛ to the velocity of human 

transportation (p. 18); yet, the example was only mentioned in passing and the emphasis on 

transportation suggests a concern with engineering rather than theoretical physics. Moreover, 

his most powerful example came from epidemiology. Hart reminded us that reliable statistics 

about infant death rates were critical in establishing that many infant deaths were the result of 

contaminated water rather than ‚summer complaint‛ or ‚the will of Heaven,‛ and that the 

knowledge derived from such statistics allowed the mortality rate to be dramatically reduced 

(p. 17). In brief, though Hart mentioned tools like thermometers and disciplines like 

aerodynamics, he was primarily interested in replicating the successes of applied sciences like 

epidemiology than of theoretical branches of physics and other natural sciences.  

The main impetus behind measures of happiness and satisfaction appears to have been a 

moral impulse to understand and improve society, specifically, an ambition to identify the 

causal antecedents of happiness in order to build a happier, and therefore better, society. The 

ambition is plainly evident throughout the history of measures of happiness and satisfaction. In 

early literature, authors like Davis, Hamilton, and Terman were clearly trying to identify the 

factors that made for a happy marriage because they thought that a grasp of these factors 

would allow them to promote success in marriage. Terman said that he was not committed to 

the view that ‚personal happiness is the only ‘proper’ goal of marriage,‛ but then added: ‚It is 

of the very nature of happiness that, other things being equal, it should be preferred to its 

opposite‛ (1938, p. 2). And Burgess and Cottrell, as we saw above, believed that marital 

adjustment and incompatibility had become a social problem and therefore of public concern 

(1939, p. 1).  

Educational psychologists like Watson and Hartmann shared the ambition. As we saw in 

Section 2.2, Watson said he studied happiness because no other human quest had a larger 

following, because even ‚the highest ethics‛ emphasized the importance of human happiness, 

and because educational programs were justified only in terms of their contribution to 

happiness (1930, p. 79). Unlike some of the later authors, Watson did not frame his project in 

terms of wellbeing or welfare. Yet, his use of the term ‚highest ethics‛ suggests that his desire 



The evolution of eupathics 

Angner 

 

www.internationaljournalofwellbeing.org 23 

to explore scientifically the determinants and distribution of happiness may have been due to 

philosophical or religious commitments to happiness as a central constituent of the good life.  

Hartmann (1934) started out by quoting Raymond Dodge, past president of the American 

Psychological Association. In a brief autobiographical narrative published in 1930, Dodge had 

written:  

I am desirous of participating in the solution of < the problem of protracted 

human happiness. Whether or not it fits in with one’s philosophy of life, the fact 

is incontestable that happiness is an important if not the most important aim of 

human endeavor (Dodge, 1930, p. 119; quoted in Hartmann, 1934, p. 202).3  

Hartmann obviously agreed: ‚If happiness be one of the major goals of living, if not the only 

consciously acceptable end of life itself,‛ he wrote, ‚surely an analysis of the conditions 

fostering or hindering its attainment is an intellectual obligation of the first order‛ (1934, p. 

203). Evidently, both Watson and Hartmann assumed that a happier world would be a better 

world and wanted to understand happiness in order to promote it.  

Hartmann (1934, p. 203) made favorable mention of Abraham Myerson’s program of 

eupathics (Myerson, 1917). Myerson was Clinical Director at the Taunton State Hospital in 

Massachusetts and a Harvard neuropathologist. His goal was to establish a field of ‚Mental 

Hygiene‛ that had ‚for its aim the well being of the normal‛ (Myerson, 1917, p. 344). He called the 

program the ‚more gracious sister‛ of eugenics, which ‚largely simmers down to a program for 

the elimination of the unfit‛ (p. 344). Myerson had great hopes for the field. He appears to have 

equated wellbeing with happiness, and believed that ‚Mood‛ was an important determinant of 

happiness. Furthermore, he claimed that Mood ‚can be reached and elevated in a perfectly 

definite manner‛ (p. 344). Myerson invested eupathics with a great deal of importance. After 

asking rhetorically what was left out of the field, he answered: ‚nothing of consequence‛ (p. 

346).  

Hart and Goldings were equally explicit about their desire to better understand happiness 

in order to promote it. As we saw in Section 2.4, Hart believed that ‚*one+ of the basic purposes 

of mankind is to be happy‛ (1940, p. 16). His Euphorimeter was presented as designed ‚to 

point the way, not to administration of drugs, but to constructive measures which may relieve 

maladjustments, promote the cure of mental suffering, and open the way toward more joyous 

living‛ (p. v). The aim was to help people ‚live joyously within a menacing world < in spite of 

threats and pressures of war, of economic disaster, of our own incurable physical handicaps 

and past emotional wounds, and of the original natures of the people with whom we have to 

live and work‛ (p. 6). Similarly, Goldings explained his interest in happiness as follows:  

‚How to gain, how to keep, how to recover happiness is, in fact, for all men at all 

times the secret motive of all they do, and of all they are willing to endure.‛ This 

pithy statement by William James reflects the almost unanimous judgment of 

prescientific psychologists from Aristotle down. Most laymen, caviling aside, 

would probably agree with it (Goldings, 1954, p. 30). 

Both Hart and Goldings appear to have operated on the assumption that the scientific study of 

happiness would help us build a happier, and therefore better, world.  

                                                
3 In passing, Dodge’s writings suggest that he found inspiration less from natural science than from philosophy. In 

the same autobiographical remarks, he wrote: ‚I am inclined to believe that the study of the history of philosophy 

and the logic of science form a valuable background for estimating the possibilities and the limitations of 

experimental evidence‛ (Dodge, 1930, pp. 120-103).  



The evolution of eupathics 

Angner 

 

www.internationaljournalofwellbeing.org 24 

The epidemiology of mental health is premised on the assumption that a better 

understanding of the causes and distribution of mental health might allow us to promote it 

more effectively. Gurin et al. conclude their book by writing: ‚We hope that this study < 

contributes to the formulation of programs designed to make [psychiatric] help available to the 

maximum number of people‛ (Gurin et al., 1960, p. 406). Similarly, as we saw in Section 2.7, 

gerontology had always been centrally concerned with measuring and promoting the 

wellbeing of older people. Neugarten et al. (1961), for example, were dissatisfied with existing 

measures based on various criteria of success or competence, which they felt relied on dubious, 

value-laden assumptions; by developing better measures of wellbeing, they thought that they 

would be in a better position to understand and promote the wellbeing of older people.  

The social indicator movement, as we saw in Section 2.8, was motivated by a desire to 

develop accurate measures of ‚the quality of national life,‛ ‚the goodness of life,‛ and ‚life 

quality,‛ that is, ‚indicators that show how well off people actually are‛ (Andrews, 1989, p. 

401; Campbell, 1976, p. 117). Authors like Campbell et al. charged that widely used economic 

measures and social indicators were imperfect proxies for wellbeing, and maintained that ‚the 

nation must change from its fixation on goals which are basically economic to goals which are 

essentially psychological, from a concentration on being well-off to a concern with a sense of 

well-being‛ (1976, p. 1). They took the importance of happiness and satisfaction to be 

uncontroversial. As Andrews and Withey put it: ‚*It+ would seem fair to say that the 

importance of human satisfaction does not need to be defended‛ (1976, p. 10). The view that a 

society with more quality of life, wellbeing, etc., would be a better society infused all this work. 

Hence:  

The promotion of individual well-being is a central goal of virtually all modern 

societies, and of many units within them. While there are real and important 

differences of opinion – both within societies and between them – about how 

individual well-being is to be maximized, there is nearly universal agreement 

that the goal itself is a worthy one and is to be actively pursued (Andrews & 

Withey, 1976, pp. 6-7). 

From the outset, these authors very much wanted their work to inform public policy, as when 

Campbell et al. said they wished to provide ‚a fuller and truer representation of the state of 

society to those people who are responsible for social decisions‛ (1976, p. 5). Cantril made it 

clear that his research was ultimately aimed not only at advancing scientific understanding, but 

also at designing new institutions: ‚More reliable predictions of what people want or do not 

want, believe or do not believe, will accept or will not accept, should aid the process of creating 

new forms of economic, social, and political institutions‛ (1965, p. 3).  

The moral impulse to build a happier and therefore better society is no less evident in 

contemporary writings on subjective measures of wellbeing. As part of their discussion of 

national wellbeing accounts, Ed Diener and Seligman write:  

The most important contribution of a national system of well-being indicators 

would be that they could focus the attention of policymakers and the public 

specifically on well-being, and not simply on the production of goods and 

services; one of the main benefits of well-being measures is that they add a 

valuable perspective beyond a cost-benefit market analysis in evaluating societal 

structures and interventions (2004, p. 21).  

The hope is that national wellbeing accounts will help policymakers understand and promote 

that which really matters, which in their view is subjective wellbeing. The underlying 



The evolution of eupathics 

Angner 

 

www.internationaljournalofwellbeing.org 25 

assumption is that a happier society would be a better society. Similarly, Richard Layard writes 

that ‚*Bentham+ proposed that all laws and all actions should aim at producing the greatest 

possible happiness< I believe that Bentham’s idea was right and that we should fearlessly 

adopt it and apply it to our lives‛ (2005, pp. 111-112). The central importance of happiness for a 

good society is thought to entail that public policy should be designed to promote it. Thus, 

Diener and Seligman ‚propose that *subjective+ well-being ought to be the ultimate goal 

around which economic, health, and social policies are built‛ (2004, pp. 1-2). In these passages, 

by the way, there is no trace of physics envy.  

In sum, the history of the measurement of happiness and satisfaction confirms the thesis 

that measures of happiness and satisfaction emerged in applied branches of psychology, 

sociology, and medicine. Though happiness and the like have begun to attract attention from 

more theoretical branches of psychology, economics, and neuroscience, measures of happiness 

did not emerge there. The history also confirms that measures of happiness and satisfaction did 

not appear as a result of physics envy; while Hart mentioned tools like thermometers and 

electrocardiographs, as well as disciplines like aerodynamics, he was interested primarily in 

practical applications, and he was most impressed by the successes of quantification in applied 

disciplines like epidemiology. The drive toward quantification was due to a desire to 

understand the causal antecedents of happiness in order to build a happier and therefore better 

society.  

 

3.2 Subjective measures as a means to overcome deficiencies in human judgment  

Porter has also argued that quantification was intended to make up for perceived deficiencies 

in unaided human judgment, in particular, a lack of trust and suspicions of arbitrariness (Levy, 

2001, p. 724; Porter, 1995, p. 199). This contention helps explain why the drive toward 

quantification was strongest in ‘soft’ disciplines like psychology in general and their applied 

subdisciplines in particular. In this view, practitioners in soft disciplines were most insecure, 

and therefore most susceptible to outside and inside pressure; they turned to empirical research 

to establish answers that could not be accused of being biased by personal, religious, and other 

considerations (Porter, 1995, p. 200). The history of measurement of happiness and satisfaction 

confirms Porter’s contention: throughout the history reviewed here, scientific measurement 

was seen as a means to overcome the limitations associated with unaided human judgment, in 

particular, a lack of trust and suspicions of arbitrariness.  

Since the very early days of happiness measurement, contributors commonly complained 

that science had failed to give adequate attention to the determinants and distribution of 

happiness and satisfaction. Watson’s remark that it was ‚extraordinary almost beyond belief 

that so few attempts have been made to apply the techniques of psychological study to the 

understanding of happiness‛ (Watson, 1930, p. 79) has echoes throughout this history. For 

example, the complaint appears in Dodge’s autobiographical sketch. Notwithstanding the 

importance of happiness, Dodge wrote, ‚it has received no commensurate scientific atention 

[sic]. The theory of the happy life remains at about the level where the Greek philosophers left 

it‛ (Dodge, 1930, p. 119; cited in Wilson, 1967, p. 302). He continued:  

Scientific [data] as to the fundamental positive conditions of protracted 

happiness are conspicuous for their absence< The positive conditions of 

happiness are left largely to accident, such as the satisfaction of instinctive wants 

with its tragic disillusionments and negative adaptions, the economic pressure 

to provide a market for manufactured products, the exigencies of the labor 

market, the desire to amass wealth, or the Bolshevistic abolition of private 



The evolution of eupathics 

Angner 

 

www.internationaljournalofwellbeing.org 26 

wealth. There are numerous wise protests that protracted happiness is achieved 

by none of these things; but positive, scientific data on the real conditions are as 

inconspicuous as scientific interest in the problem (Dodge, 1930, p. 120).  

Dodge believed that a scientific approach based on scientific measurement was more promising 

than alternative approaches; that in the absence of scientific data, ‚wise protests‛ were 

distrusted and largely impotent; and that using unaided human judgment rather than ‚the 

techniques of psychological study‛ was akin to leaving it to accident. The fear of arbitrariness is 

highly evident.  

The same concern appears in Hart and Goldings’s work. Hart noted:  

By measuring accurately, scientists have trebled the length of human life – but 

then we discover that they have left a lot of people wondering whether life is 

worth living< What our scientists have been doing has been to use 

measurement to increase things which people thought would contribute to 

happiness, but without ever measuring happiness itself to see whether the 

alleged means were actually contributing to the desired end (1940, pp. 18-19).  

Quite obviously, Hart did not trust unaided human judgment to reveal what ‚actually‛ 

contributes to happiness. As we saw in Section 2.4, Hart favored a scientific approach. He 

asked: ‚Can recent advances in scientific thinking tell us more and more effectively what to do 

in order to be happy, and in order to help make our fellow human beings happy?‛ (1940, p. 16). 

He answered:  

The present book is based on two propositions: first, that it is possible to 

measure happiness and unhappiness reliably; and second, that, if we can thus 

measure, we can then move on toward discovering the causal factors by means 

of which we can learn with more and more effectiveness to eliminate misery and 

increase joy scientifically (p. 16).  

Hart believed that a scientific approach to identifying the causes of happiness, and a scientific 

approach only, would permit happiness scholars to be as effective in enhancing happiness as 

epidemiology had been in eliminating disease and aerodynamics in building airplanes. Having 

pointed to the enormous success of measurement in fields like epidemiology, he added: ‚Why 

not begin to apply these methods of precise measurement when we come down to the crucial 

question of happiness?‛ (p. 18). Goldings (1954) concurred. In spite of the virtual consensus on 

the importance of happiness, he wrote, ‚there has been a marked reluctance on the part of 

present-day researchers to undertake systematic studies of happiness, unhappiness, and related 

phenomena‛ (p. 30). When he proceeded to talk about the questions that have to be addressed 

by ‚a psychology of happiness,‛ he clearly assumed that such a psychology would uncover 

insights that would be unavailable in the absence of scientific study.  

As we saw in Section 2.8, authors like Andrews and Campbell were highly critical of 

economic measures and social indicators because of our ignorance of how objective conditions 

translated into subjective experience, which, they thought, was what ultimately mattered. As 

Campbell et al. wrote, ‚the fact is that we do not know how well objective measures like these 

represent underlying psychological states or how well social indicators can be taken to 

represent the quality of life experience‛ (1976, p. 3). Their research aimed to make up for this 

deficiency:  

The research with which this book is concerned derives from the conviction that 

the relationship between objective conditions and psychological states is very 

imperfect and that in order to know the quality of life experience it will be 



The evolution of eupathics 

Angner 

 

www.internationaljournalofwellbeing.org 27 

necessary to go directly to the individual himself for his description of how life 

feels to him. This obviously will take us into the subjective world of perceptions, 

expectations, feelings, and values and will involve us in excruciating problems 

of definition and measurement (p. 4).  

While Campbell and his coauthors recognized that a scientific study of subjective phenomena 

would face difficult problems, they clearly believed that a non-scientific approach would be 

even less promising.  

The theme is echoed in contemporary writings as well. Seligman maintains that when it 

comes to what makes us happy, ‚the field of psychology *is+ a puzzling disappointment‛ (2002, 

p. ix). And Diener and Seligman target public policy when they write:  

While wealth has trebled over the past 50 years, for example, well-being has 

been flat, mental illness has increased at an even more rapid rate, and data, not 

just nostalgic reminiscences, indicate that the social fabric is more frayed than it 

was in leaner times. These inadequacies lead us to advocate that an ongoing 

system of indicators be instituted by governments and organizations to track 

well-being over time. It is clear that policymakers now care about well-being, in 

addition to economics, because policies are being created on the basis of mere 

guesses and romantic sentiments about what will enhance well-being (e.g., 

family leave). These guesses are undoubtedly correct in some cases, but they are 

incorrect in other cases. For this reason, ongoing measurement of well-being in 

representative samples and in diverse domains of life is required to confirm or 

disconfirm the efficacy of policies instituted to increase well-being (2004, pp. 20-

21).  

In Diener and Seligman’s view, the problem can be solved by shifting from unaided human 

judgment to a system of national wellbeing indicators, which would provide the hard data 

required to eliminate the reliance on unscientific beliefs. Interestingly, proponents of subjective 

measures have also turned the tables on their critics, who do not trust the validity and 

reliability of subjective measures, and attacked them for failing to marshal any empirical 

evidence that undercuts the validity of such measures (Diener, Lucas, Schimmack, & Helliwell, 

2009, p. 75). This move, triggered by the charge of arbitrariness and the lack of trust, implicitly 

faults the critics of subjective measures with relying too heavily on unaided human judgment.  

In brief, the history of happiness and satisfaction measurement confirms the contention that 

scientific measurement was seen as a means to overcome the limitations associated with 

unaided human judgment, in particular, a lack of trust and suspicions of arbitrariness. 

Operating on the assumption that a happier society would be a better society and that 

happiness therefore is worth pursuing (cf. Section 3.1), these researchers doubted that unaided 

human judgment was sufficient to identify the antecedents of happiness and considered a 

scientific approach more likely to accomplish this goal. The concern was that hypotheses 

generated by non-scientists were no better than mere guesses and romantic sentiments and that 

acting on such hypotheses was tantamount to leaving it to chance. The contention that 

subjective measures were seen as a means to overcome the limitations associated with unaided 

human judgment is consistent with the fact that the drive toward measuring happiness and the 

like appeared in applied subdisciplines like educational psychology, which are most frequently 

criticized for being ‘soft’ and therefore most vulnerable to a lack of trust and suspicions of 

arbitrariness.  

 



The evolution of eupathics 

Angner 

 

www.internationaljournalofwellbeing.org 28 

3.3 Disagreements about the nature of wellbeing 

Porter has also proposed that the drive toward measurement is not typically impeded by 

disagreements about the nature of the object under study (M. S. Morgan, 2001, p. 248; Porter, 

1995, pp. 94-95). The point was made in the context of attitude measurement as early as 1935, 

when Allport noted that ‚attitudes today are measured more successfully than they are 

defined‛ (Allport, 1935, p. 828). The history of happiness and satisfaction measurement 

confirms Porter’s proposition: the researchers discussed here (insofar as they expressed a view 

at all) had radically different ideas about the nature of wellbeing, happiness, and satisfaction. 

Yet, the disagreement was not treated as an obstacle to measurement.4  

Students of marital success, as we saw in Section 2.1, were deeply interested in happiness, 

satisfaction, and the like, though their focus frequently was on the happiness of the marriage as 

opposed to that of the people in it. These figures had little to say about the nature of happiness, 

satisfaction, and success, though the terms appear to be treated as interchangeable, e.g., when 

Davis asked whether people considered their lives ‚happy, satisfactory, successful‛ or 

‚unhappy, unsatisfactory, unsuccessful‛ (1929, p. 89).  Educational psychologists were no less 

interested in happiness and satisfaction, though their focus was squarely on individuals rather 

than on marriages. They too appear to have treated happiness and satisfaction as 

interchangeable; when Watson (1930) talked about satisfaction as well as happiness – e.g., in 

the first sentence of his paper – there is no evidence that he used the terms in any other way 

than as synonyms. He did say: ‚What is studied might, in the strictest sense be termed not 

happiness, but self-estimates of happiness‛ (p. 79). Yet, the fact that from that point on he 

talked about happiness suggests that in Watson’s mind the two were at least tightly linked.  

Terman (1938) eschewed definitions. In order to avoid ‚philosophical connotations,‛ he 

wrote, he ‚preferred to apply the term in a sense familiar to everyone‛ (p. 3) without specifying 

what that sense was. Terman immediately went on to offer what appears to be a purely 

operational definition: ‚The important thing to remember is that throughout this discussion 

happiness or unhappiness will mean a high or low score on a particular scale‛ (p. 4). The scale 

in question was constructed by combining subjective ratings with other ‚more objective‛ 

judgments, but from the weighting scheme it is clear that he considered the subjective rating 

the most important item. Sometimes, Terman (e.g. on p. 367) referred to his marital happiness 

score as an index of satisfaction, which suggests that he treated ‚happiness‛ and ‚satisfaction‛ 

as interchangeable.  

Burgess and Cottrell relied on Webster’s New International Dictionary, which defined 

‚happiness‛ as ‚a state of well-being characterized by relative permanence, by dominantly 

agreeable emotion ranging in value from mere contentment to positive felicity, and by a natural 

desire for its continuation‛ (Burgess & Cottrell, 1939, p. 31). As the authors note, the dictionary 

definition emphasizes the subjective nature of the concept. Interestingly, they made no attempt 

at explaining the meaning of the word to their subjects, on the assumption that people in 

general would understand the notion in accordance with the dictionary definition (p. 31). Like 

Terman, Burgess and Cottrell also sometimes talked about ‚satisfaction‛ and ‚success‛ instead 

of ‚happiness‛ (p. 45).  

Hart (1940) appears to have been somewhat ambivalent, as he offered two different 

definitions of ‚happiness.‛ According to the first: ‚Happiness is the state in which people are 

when they say sincerely, ‘I am happy,’ and it is the opposite of the state in which they are when 

                                                
4 See Angner (2010) for a more careful review of accounts of well-being implicit in the literature on subjective 

measures.  



The evolution of eupathics 

Angner 

 

www.internationaljournalofwellbeing.org 29 

they say sincerely, ‘I am unhappy’‛ (p. 182). He added: ‚It has been assumed that the average 

person is sincere in his answers to such a test as the Euphorimeter, especially when he does not 

sign his name < Moreover, it has seemed reasonable to assume that whatever insincerities do 

occur tend to cancel each other out‛ (p. 182). He noted that the first definition, under the 

assumption of sincerity, ‚serves operationally as a basis for measurement, but it is rather 

deficient in providing insight,‛ and ‚does not take us very far into the meaning of the term‛ (p. 

183). Thus, he proposed a second definition: ‚Happiness is any state of consciousness which 

the person tested seeks to attain or to maintain, and it is the opposite of any state which the 

possessor seeks to change or from which he seeks to escape or withdraw‛ (p. 183). In the end, 

however, Hart argued that ‚the two definitions really define essentially the same thing‛ (p. 

184). The name of the Euphorimeter suggests that Hart also identified happiness and euphoria.  

Like Hart, Goldings was unusually explicit about ‚the intrinsic nature and characteristics‛ 

of happiness. He wrote: ‚Happiness and unhappiness may be considered as zones on a 

continuum of hedonic affect which embraces feelings of elation, contentment, satisfaction, and 

pleasure at the positive pole and feelings of depression, discontent, and unpleasure at the 

negative pole‛ (Goldings, 1954, p. 31). He added: 

*Although+ in common parlance, ‚happiness‛ refers only to the positive 

elements of hedonic feeling, in the present investigations ‚happiness‛ more 

generally refers to the entire continuum as a whole and ‚satisfaction‛ and 

‚dissatisfaction‛ refer more specifically to the polar regions of positive and 

negative hedonic feeling (p. 31).  

From this quote, it appears that Goldings drew a sharp distinction between ‚happiness‛ as the 

continuum of hedonic feeling and ‚satisfaction‛ as the positive end of the continuum. It is 

unclear whether he in fact used the terms in this way, however. Consider:  

In order to bring happiness and unhappiness (or satisfaction and dissatisfaction) 

into proper perspective vis-a-vis other general areas of psychological study, they 

may be considered hypothetical physiological states with both subjective 

manifestations (diffuse and pervasive feelings of pleasantness and satisfaction, 

and of unpleasantness and dissatisfaction) and objective signs (posture, gait, facial 

expression, tone of voice; feelingful verbal statements; indirect or projective 

indices of feeling) (p. 31).  

Here, happiness and satisfaction were both treated as physiological states that cause subjective 

feelings as well as outward behaviors of a certain kind.  

Campbell and colleagues (1976) differed from their predecessors by drawing a sharp 

distinction between ‚happiness‛ and ‚satisfaction.‛ Having asked questions about satisfaction 

alongside questions of happiness, Campbell et al. realized that answers to questions of 

satisfaction were not as highly correlated with answers to questions about happiness as they 

would have thought (p. 8). They wrote: ‚This means that there is some significant minority of 

persons who report relative happiness along with relative dissatisfaction with their lives, and 

the converse‛ (p. 9). As a result, they concluded that happiness and satisfaction were two 

different traits, both of which were constitutive of wellbeing, and which had to be measured 

independently. ‚Satisfaction implies a judgmental or cognitive experience,‛ they maintained 

‚while happiness suggests an experience of feeling or affect‛ (p. 8). The weak correlation 

between happiness and satisfaction, they continued, also means that the choice of measure may 

affect the conclusions of a study. Campbell et al. favored a measure of satisfaction – as opposed 

to happiness – as a measure of the quality of life experience. They offered several  reasons. First, 



The evolution of eupathics 

Angner 

 

www.internationaljournalofwellbeing.org 30 

they took ‚satisfaction‛ to be easier to define and to translate than ‚happiness‛ (pp. 7-8). 

Second, they wanted their data to be relevant to public policy, and they thought legislators and 

decision-makers were more accustomed to thinking in terms of satisfying needs than of 

promoting happiness (pp. 8-9). Third, Campbell et al. wanted a concept appropriate to use both 

in the context of ‚life as a whole‛ as well as in the context of more limited domains (p. 9).  

Bradburn and Caplovitz (1965) made another unexpected discovery, viz., that positive 

affect was remarkably weakly correlated with negative affect. In order to develop what they 

called ‚more detailed measures of well-being,‛ and using a strategy reminiscent of Watson’s 

adjective-based measure, Bradburn and Caplovitz offered twelve different descriptions, half of 

them positive, half negative, and asked subjects how often they had felt that way during the 

past week (pp. 15-16). They wrote: ‚We expected that the items would cluster in two groups, 

one indicative of positive and the other of negative feelings, and that the two clusters would be 

negatively related to one another‛ (p. 16). As expected, the authors did find ‚a strong tendency 

for most of the items to fall into two clusters of positive and negative feelings‛ (p. 16). 

However, they were surprised to learn that ‚items in one cluster are not negatively related to 

those in the other cluster in any consistent or strong fashion‛ (p. 18). The two items which were 

the most strongly (negatively) correlated were ‚depressed or very unhappy‛ and ‚on top of the 

world,‛ and these items had a correlation of a mere -.19 (p. 18). Interestingly, while positive and 

negative items did not correlate with each other, each was correlated with happiness (pp. 18-

19). The authors inferred that ‚happiness is a result of the relative strengths of positive and 

negative feelings, rather than of the absolute amount of one or the other‛ (p. 21). In light of this 

finding, it is sometimes suggested that wellbeing should be seen has having three irreducible 

components: the presence of positive affect, the absence of negative affect, and satisfaction.  

In sum, the history of the measurement of happiness and satisfaction confirms Porter’s 

proposition that the drive toward measurement is not typically impeded by disagreements 

about the nature of the object under study: the researchers discussed here (insofar as they 

expressed a view at all) had radically different ideas about the nature of happiness and 

satisfaction, but the disagreement was not treated as an obstacle to measurement. Many 

researchers in this tradition declined to discuss what they meant by ‚happiness‛ and 

‚satisfaction.‛ Some of them simply noted that they use the concept in the ‚established‛ sense, 

and that they assumed their subjects to do so too; some made it explicit that they hoped to 

avoid philosophical discussion. When authors did offer explicit definitions of ‚happiness,‛ 

‚satisfaction,‛ and ‚well-being,‛ not only did they disagree with each other regarding the 

proper understanding of these terms, but there are passages in which some (like Hart and 

Goldings) appear unable to agree with themselves. Moreover, the perceived relationship 

between happiness and satisfaction shifted during the period under study. While the early 

contributors treated happiness and satisfaction as largely interchangeable, Campbell et al. 

thought of them as different and of wellbeing as constituted by both happiness and satisfaction, 

and Bradburn and Caplovitz’s results have inspired the view that wellbeing consists of three 

irreducible components: the presence of positive affect, the absence of negative affect, and 

satisfaction. Yet, the highly evident disagreement about the nature of happiness was not treated 

as an obstacle to measurement.  

 

3.4 Convergence in measurement  

In light of the disagreement about the nature of the object under study, it is interesting to note 

that, at least during the episode surveyed in this paper, there was a certain degree of 

convergence regarding how to measure happiness and satisfaction. It is true that since the 



The evolution of eupathics 

Angner 

 

www.internationaljournalofwellbeing.org 31 

1920s and 30s a variety of instruments have been used to gauge people’s happiness and 

satisfaction; at no point was there consensus about the proper way to measure wellbeing. Yet, 

there was a clear tendency to substitute a direct question like that of Gurin et al. (1960) for more 

complex questionnaires. In addition, some found it more convenient to use preexisting 

measures than to develop their own. Consequently, in time, there was considerably more 

agreement about how to measure wellbeing than about how to define it.  

Watson (1930), as pointed out in Section 2.2, used a surprising variety of remarkably 

sophisticated measures to assess his subjects’ happiness. First, he asked: ‚Comparing yourself 

with other persons of the same age and sex how do you feel you should rate your own general 

happiness?‛ (p. 79). The subjects were then given a horizontal line, with captions that read, 

from left to right, ‚Most miserable of all,‛ ‚About three-fourths of the population happier than 

you are,‛ ‚The average person of your own age and sex,‛ and so on (p. 80). Subjects were asked 

to make a short vertical mark at the point where they thought they belonged, taking into 

account their ‚average state over several months‛ (p. 80). The subjects were also asked to place 

a circle at the point where they thought their friends would rate them (p. 80). Second, subjects 

were offered ten descriptions, of one or two sentences each, and asked: ‚Among the following 

descriptions, arranged in miscellaneous order, choose the one which comes nearest to fitting 

you‛ (p. 80). The descriptions included the following: ‚Cheerful, gay spirits most of the time. 

Occasionally bothered by something but can usually laugh it off,‛ ‚Ups and downs, now 

happy about things, now depressed. About balanced in the long run,‛ and ‚Life often seems so 

worthless that there is little to keep one going. Nothing matters very much, there has been so 

much of hurt that laughter would be empty mockery‛ (p. 81). Third, subjects were given a 

blank space and asked: ‚Now write in a sentence or two, something like those above, which 

you believe will most truly describe your own general happiness in life‛ (p. 81). Fourth, 

subjects were given a list of fifty properties – ‚Enthusiastic,‛ ‚Troubled,‛ ‚Annoyed,‛ and so 

on – and asked: ‚Check every term which you believe could fairly be applied to yourself in 

your prevalent attitudes‛ (p. 81). Half of the descriptions were positive, half negative (p. 81). 

Finally, using a graphic rating device as in the first question, subjects were asked about their 

happiness in different areas and in different stages of life (p. 82).  

On the basis of subjects’ responses, Watson computed an aggregate ‚happiness score‛ (p. 

82). To check for reliability, Watson used the responses to different items to create two different 

happiness scores. He noted that the correlation between the scores for men and women were as 

high as .83 and .85 respectively. As a result of these calculations, he concluded: ‚This indicates 

that the measure is reasonably consistent, throughout‛ (p. 83). Watson also noted that the 

correlation coefficient between the subjects’ own rating of their happiness on a graphical scale 

and the aggregate happiness scores for men and women were .81 and .82 respectively (p. 86). 

Thus, he inferred that for some purposes, self-ratings may suffice. He wrote: ‚For the purpose 

of separating high and low groups such a simple indicator would probably be adequate‛ (p. 

86). Based on his own results, Sailer drew the same conclusion: ‚In general, self-estimate of 

happiness proved to be a characteristic easily measured and showing considerable consistency 

and relationship to many other sorts of factors‛ (1931, p. 100).  

Hart (1940) also used a range of measures. For his ‚At-the-Moment Euphorimeter,‛ Hart 

modified one of the kinds of test used by Watson (1930) and Sailer (1931). Participants were 

presented with a list of adjectives and asked to underline each adjective ‚which fairly well 

describes the way you have felt a good deal of the time or several times since you woke up this 

morning‛ and to cross out each adjective ‚which does NOT describe the way you have felt at 

any time today‛ (Hart, 1940, p. 114). The adjectives were chosen by picking all synonyms and 



The evolution of eupathics 

Angner 

 

www.internationaljournalofwellbeing.org 32 

antonyms of ‚happy‛ and ‚unhappy‛ in a dictionary, then adding all synonyms and antonyms 

of those, and so on, until he had a list of 48 adjectives (pp. 176-177). A subject was said to have a 

happy reaction each time he or she underlined a positive adjective or crossed out a negative 

one, and an unhappy reaction each time he or she underlined a negative adjective or crossed 

out a positive one (pp. 115-116). The test score could then be computed based on the number of 

happy and unhappy reactions (p. 177). After administering the test to 2,200 subjects, Hart 

assured himself that the test was reliable by computing scores separately for each half of the 

adjectives and finding the reliability of the entire test to be .90 (pp. 178-179). For the ‚Long-Run 

Euphorimeter‛ Hart borrowed a number of questions from earlier studies and added some 

‚which seemed likely to be valuable‛ (p. 181). Many of the questions asked subjects how they 

‚usually‛ felt, rather than how they felt at the moment (p. 111). The scoring system for the 

Long-Run Euphorimeter was calibrated by comparing the results of 2,200 subjects who took 

both the Long-Run and the At-the-Moment test (p. 181).  

The validity of his measures was assured, Hart believed, by the manner in which he had 

defined ‚happiness.‛ In effect, he argued, ‚the adjective battery consists in offering an 

opportunity to say ‘I am happy’ in 48 different ways, or to say ‘I am unhappy’ in 48 different 

ways, or to make whatever compromise between these extremes fit one’s own state at the time 

tested‛ (p. 182). The Long-Run test also contained questions concerning how subjects felt about 

the change or maintenance of current conditions. Hart wrote:  

At every crucial point in a long series of statistical analyses, based on these data, 

we have found that persons that say, in one form or another, ‚I am happy,‛ tend 

also to say, in various ways, ‚I want to keep my present way of life developing 

as it is now going‛ < The correlation is so close that it seems wisest to proceed 

upon the assumption that our two definitions of happiness really define the 

same thing (p. 184).  

He concluded:  

Various improvements need to be made, and will continue to be made, in the 

Euphorimeter tests. But it is believed that, as they now stand, with their present 

scoring methods, the Euphorimeters measure happiness and unhappiness with 

sufficient reliability and validity to identify outstandingly happy and unhappy 

people and to give important aid in the process of increasing the happiness of 

those tested (pp. 184-185).  

Hence, Hart convinced himself on empirical grounds that the two definitions of happiness 

were interchangeable, and that the two Euphorimeters – both of which were based on various 

kinds of self-reports – generated measures of happiness that were reliable and valid.  

Goldings participated in a study on ‚personality development and projective techniques‛ 

(1954, p. 30), and as we saw, developed some original methods to assess happiness. First, his 

subjects were shown 30 photographs of faces with ambiguous facial expressions, and asked ‚to 

rate the satisfaction or dissatisfaction (happiness or unhappiness) of the people‛ (p. 34). He 

added: ‚The Ss [subjects] rated each picture twice on a 10 point scale (extremely unhappy to 

extremely happy). The sum of the 60 ratings (2 ratings for each of the 30 pictures) constituted 

the S’s ascribed-happiness score‛ (p. 34). The reason why Goldings included this test was the idea 

that ‚direct avowal of happiness may tend to be subjected to some distortion,‛ and that it 

would be important to have alternative, indirect means of measuring happiness. The 

significance of the happiness ascribed to others with ambiguous facial expressions is that the 

subject would be expected, on theoretical grounds, to project his or her own happiness onto the 



The evolution of eupathics 

Angner 

 

www.internationaljournalofwellbeing.org 33 

faces (p. 33). Second, subjects were given linear-type rating scales like those used by Watson 

(1930) and Sailer (1931) (Goldings, 1954, pp. 34-35). On the basis of the results, Goldings 

computed an ‚avowed happiness‛ score ranging from 1 to 6 (p. 35). Third, Goldings asked five 

experimenters to rank-order the subjects ‚on ‘general happiness and overall satisfaction’ based 

on the experimenter’s subjective, clinical judgment of the individual‛; the ratings were 

retrospective, and based on the impression made by the subjects during previous testing 

periods (p. 35). 

Goldings found that there was ‚highly significant agreement among the five judges,‛ and 

that there was ‚a fairly close general agreement between the S’s avowal of his own happiness 

and the ratings assigned by the clinical judges‛ (1954, p. 40). In contrast, he found no positive 

correlation between avowed happiness and projected happiness. A closer examination of the 

data suggested to Goldings that those subjects whose avowed happiness scores were in the 

normal range indeed tended to attribute their own mental state to the ambiguous faces, 

whereas those subjects who were extremely happy or unhappy tended to attribute the opposite 

state (p. 42). Hence, he dismissed the indirect measure as a measure of happiness, noting that 

‚the general hypothesis of a general supplementary projection of happiness must be rejected‛ 

(p. 41).  

Bradburn and Caplovitz (1965) agreed with Goldings (1954) that self-report measures were 

no less adequate than any other kind of measure. In their view, the fact that no other measure 

had been shown to be decidedly more valid or reliable, entailed that psychological wellbeing 

may as well be assessed by simple self-report measures. They wrote:  

[There] is no evidence that self-reports are any less (or for that matter more) 

valid than expert ratings or psychological tests for rating people on a mental 

health dimension. Furthermore, self-reports have the eminently practical virtues 

of face validity, directness, and ease of use (Bradburn & Caplovitz, 1965, p. 7).  

The authors were aware, of course, that there was no agreement on what to call the dimension 

of interest (p. 1). Yet, in light of all the evidence, they could identify no reason to favor more 

complex measures over self-reports. In practice, Bradburn and Caplovitz relied heavily on the 

question from Gurin et al. (1960).  

Neugarten et al. constructed two indices of life satisfaction of their own. The first (Life 

Satisfaction Index A) consisted of 20 statements of the form ‚I’ve gotten pretty much what I 

expected out of life‛ and checkboxes for ‚AGREE,‛ ‚DISAGREE,‛ and ‚UNSURE‛ (1961, p. 141). 

The score was computed by awarding the respondent one point every time he or she agreed 

with a statement indicating satisfaction or disagreed with one indicating dissatisfaction. The 

second index (Life Satisfaction Index B) consisted of twelve questions such as ‚How satisfied 

would you say you are with your way of life?‛ and three options ‚very satisfied‛ (for two 

points), ‚fairly satisfied‛ (one point), and ‚not very satisfied‛ (zero points) (p. 142).  

When confronted with Cantril’s ‚Self-Anchoring Striving Scale‛, the subject was asked: 

‚When you think about what really matters in your own life, what are your wishes and hopes 

for the future? In other words, if you imagine your future in the best possible light, what would 

your life look like then, if you are to be happy?,‛ and similarly for the worst possible life (1965, 

p. 23). Thereafter, the subject was shown a non-verbal device referred to as ‚the ladder of life,‛ 

a drawing of a ladder-like shape where the rungs had been numbered from 0 to 10 (p. 22). 

Next, the subject was told: ‚Here is a picture of a ladder. Suppose we say that the top of the 

ladder (POINTING) represents the best possible life for you and the bottom (POINTING) 

represents the worst possible life for you,‛ and asked ‚Where on the ladder (MOVING FINGER 

RAPIDLY UP AND DOWN THE LADDER) do you feel you personally stand at the present time?‛ (p. 



The evolution of eupathics 

Angner 

 

www.internationaljournalofwellbeing.org 34 

23).5 Cantril’s basic idea was that this device would allow him to judge a person’s 

circumstances from the person’s own point of view. In this test, Cantril wrote: ‚A person is 

asked to define on the basis of his own assumptions, perceptions, goals, and values the two 

extremes or anchoring points of the spectrum on which some scale measurement is desired‛ (p. 

22). To his surprise, Cantril learned that in some cultures ladders are uncommon; thus, he 

sometimes substituted a picture of a mountain for the ladder (Andrews & Robinson, 1991, p. 

73).  

Campbell et al. (1976) asked questions about subjects’ satisfaction in specific domains – 

including marriage, job, housing, etc. – as well as about their satisfaction with ‚life in general‛ 

and about their happiness, using the question from Gurin et al. (1960). Finally, Campbell et al. 

computed what they called an ‚Index of General Affect‛ on the basis of ‚a series of descriptive 

adjectives‛ that subjects could use to describe their lives (1976, p. 13). The idea was to explore 

how domain satisfaction related to general life satisfaction, and how the latter related to other 

measures of the ‚quality of life experience.‛ For a great many of the comparisons, however, 

Campbell et al. relied on a composite ‚Index of Well-Being,‛ which was ‚a single variable 

representing a rather global sense of well-being,‛ and which combined the satisfaction score 

and the result on the adjective test (p. 49).  

Andrews and Withey (1976, Ch. 6) tested 68 different measures but ended up favoring one 

they thought had more desirable qualities than the others, viz., the Delighted-Terrible (D-T) Scale, 

which consists of seven categories ranging from ‚Delighted,‛ though ‚Pleased,‛ and ‚Mostly 

satisfied,‛ to ‚Mixed (about equally satisfied and dissatisfied)‛ to ‚Mostly dissatisfied,‛ 

‚Unhappy‛ and ‚Terrible‛ (p. 18). The subjects were given three additional response options: 

‚Neutral (neither satisfied nor dissatisfied),‛ ‚I never thought about it,‛ and ‚Does not apply to 

me‛ (pp. 18-19). Andrews and Withey offered a variety of reasons to think that people make 

the kind of global judgments that their research presupposed and that their measures were 

valid. One of the reasons they thought it was meaningful to ask for global assessments was that 

people articulate such judgments ‚promptly and with apparent ease‛ (p. 64). Moreover, in their 

sample, less than one percent checked the box marked ‚I never thought about it‛ when asked 

how they felt about their life as a whole (p. 64). Finally, the authors suggested, the fact that 

questions such as ‚How are things?‛ are so common suggests that people think in those terms 

(pp. 64-66).  

As this review makes clear, since the 1920s and 30s, a variety of instruments have been used 

to gauge people’s happiness and satisfaction; at no point was there consensus about the proper 

way to measure wellbeing. Nevertheless, at least during the episode surveyed in this paper, 

there was a certain degree of convergence regarding how to measure happiness and 

satisfaction. Many of the early contributors, like Watson, Sailer, and Hart used a remarkable 

variety of measures of happiness and satisfaction. However, over time there was a clear 

tendency to substitute a direct question like that of Gurin et al. (1960) for more complex 

questionnaires (cf. Diener et al., 1999, p. 277). Having compared the performance of a variety of 

measures, Watson, Sailer, and many others concluded that self-reports appeared to perform as 

well as any other measure and sometimes better; certain measures, such as Golding’s 

projection-based measures, appear to have been conclusively rejected. In addition, some 

contributors found it useful to employ preexisting measures rather than to develop their own. 

Among other things, as Wilson (1967, pp. 304-305) had pointed out, the use of the same tool as 

earlier researchers permits the construction of longer time series data. While some 

                                                
5 The parenthetical notes in small caps are Cantril’s instructions to the experimenter (Cantril, 1965, p. 22).  



The evolution of eupathics 

Angner 

 

www.internationaljournalofwellbeing.org 35 

disagreement remained, much of it appears to reflect a disagreement about the nature of 

wellbeing (cf. Section 3.3) rather than a disagreement about how best to measure it. Either way, 

in time, there was considerably more agreement about how to measure wellbeing than about 

how to define it.  

 

3.5 Summary and discussion  

In this section, I have argued that the history of subjective measures of wellbeing from the 

1920s or 30s, in all essentials, confirms Porter’s general account of the history of measurement 

in social and behavioral science. Consistent with Porter’s account, subjective measures of 

wellbeing emerged in applied branches of psychology, sociology, and medicine, and they 

appeared not as the result of an effort to replicate the successes of quantification in the natural 

sciences, but rather of an ambition to understand the causal antecedents of happiness and 

satisfaction in order to build a happier and therefore better society. Throughout the history 

reviewed here, scientific measurement was seen as a means to overcome the limitations 

associated with unaided human judgment, in particular, a lack of trust and suspicions of 

arbitrariness. The researchers discussed here (insofar as they expressed a view at all) had 

radically different ideas about the nature of wellbeing, happiness, and satisfaction, but the 

disagreement was not treated as an obstacle to measurement. Because of a tendency to 

substitute a direct question for more complex questionnaires, and because some found it useful 

to employ preexisting measures, in time, there was considerably more agreement about how to 

measure wellbeing than about how to define it. As a result, some of the self-report measures 

discussed here remain commonly used. This is true, for instance, in the case of the measures 

used by Andrews and Withey (1976), Cantril (1965), and Gurin et al. (1960).  

 

4. Conclusion 

In this paper I have traced the historical roots of measures of happiness, satisfaction, and the 

like. The history turned out to be longer than is typically suggested. Far from being a recent 

development, subjective measures can be traced back at least to the 1920s and 1930s. The 

appearance of a science of happiness was perhaps marked by Myerson’s 1917 study promoting 

the discipline of eupathics, the ‚more gracious sister‛ of eugenics, focusing on ‚the well being of 

the normal‛ (Myerson, 1917, p. 344; cf. Section 3.1). Systematic happiness research took off 

shortly thereafter. Subjective measures emerged in marital success studies, educational 

psychology, and personality psychology in the 1920s and 30s, as tools developed for the study 

of personality were applied to happiness and satisfaction; the measures were further shaped by 

the epidemiology of mental health, gerontology, and the social indicator movement in the 1960s 

and 70s, as measures of happiness and satisfaction were recruited as proxies for mental health 

and wellbeing in large, representative samples. The history is not only longer, but also richer, 

than typically suggested. As we have seen, the researchers in the traditions discussed here had 

rather different purposes, coming as they did from a variety of backgrounds and being 

involved in a variety of research projects. Social and behavioral scientists who studied 

happiness, satisfaction, wellbeing, and the like operated with a variety of definitions. The 

scientists invented a fascinating variety of measurement tools, some of which were ultimately 

rejected, but some of which – like the question of Gurin et al. (1960) – turned out to have a 

remarkable longevity. A number of the studies were longitudinal; some used remarkably large, 

representative samples.  

The history of the measurement of happiness and satisfaction is consistent with Porter’s 

general account of measurement in social and behavioral science. One of Porter’s main theses is 



The evolution of eupathics 

Angner 

 

www.internationaljournalofwellbeing.org 36 

that the drive toward quantification emerged in applied rather than theoretical branches of 

social science, and that it did so not as a result of physics envy, but rather of a moral impulse to 

understand and improve a changing society. Indeed, measures of happiness and satisfaction 

emerged in applied branches of psychology, sociology, and medicine, and they appeared not as 

the result of an effort to replicate the successes of quantification in the natural sciences, but 

rather of a desire to understand the causal antecedents of happiness in order to build a happier 

and therefore better society. Porter also argued that quantification often appeared as a means to 

overcome deficiencies in human judgment, in particular, a lack of trust and suspicions of 

arbitrariness. In fact, throughout the history reviewed here, scientific measurement was seen as 

a means to overcome the limitations associated with unaided human judgment, in particular, a 

lack of trust and suspicions of arbitrariness. In addition, Porter maintained that the drive 

toward measurement is not typically impeded by disagreements about the nature of the object 

under study. As it happens, the researchers discussed here (insofar as they expressed a view 

about the nature of happiness and satisfaction at all) had radically different ideas about the 

nature of wellbeing, but such disagreement was not treated as an obstacle to measurement – 

indeed, in time, there was considerably more agreement about how to measure wellbeing than 

about how to define it.  

The story confirms Chang’s thesis that history of science as complementary science can 

serve multiple purposes (2004, p. 249). First, the story offers further proof that history can serve 

to recover knowledge about natural and social phenomena as well as about previous efforts to 

measure them that appears to have been forgotten. The variety and sophistication of the 

measures used even early in the game, by researchers like Watson and Hart, is astonishing and, 

it appears, largely forgotten by practicing social and behavioral scientists. Recent work has 

noted a surprisingly weak correlation between objective conditions and subjective wellbeing 

(Argyle, 1999). But this phenomenon was well known throughout the history reviewed here; 

Symonds, for example, concluded: ‚Happy and unhappy *people+ are remarkably alike in their 

problems and interests. The unhappy do not have peculiar problems but make less satisfactory 

adjustments to their problems‛ (1937, p. 293). Meanwhile, recent years have seen a sharp 

debate about the extent to which people adapt to objective circumstances. Under the heading 

‚Myths in the science of happiness,‛ Diener criticizes what he calls the ‚extreme setpoint 

theory‛ and argues that circumstances do matter and that adaptation is not in general complete 

(Diener, 2008, pp. 494-496). But already, in 1940, Hart had shown that recently married couples, 

prisoners working on roads, and couples who have contemplated divorce exhibit sharply 

different happiness scores, which would disconfirm the extreme setpoint theory (see Figure 1).  

The story also confirms that history can serve to develop critical awareness of 

contemporary efforts, their prospects and limitations. Among other things, the historical 

overview offered here sheds light on the methodological and philosophical underpinnings of 

the project. As one example, critics of subjective measures sometimes suggest that proponents 

of subjective measures of wellbeing illegitimately assume without argument that people are 

capable of accurately reporting how happy or satisfied they are. This impression is sometimes 

encouraged by the proponents themselves. Consider again Frey and Stutzer, who state: ‚In 

general, it can be assumed that [individuals] are the best judges of when they are happy and 

when they are unhappy‛ (2002, p. 4). They support this assumption by referring to ‚a sensible 

tradition in economics to rely on the judgment of the persons directly involved‛ (p. 4). This 

move is surprising, given that only a few pages earlier they attack the track record of economics 

in saying something useful about happiness (p. vii). However, as we have seen, it is a mistake 

to think that psychologists and economists working on subjective wellbeing simply take 



The evolution of eupathics 

Angner 

 

www.internationaljournalofwellbeing.org 37 

people’s ability to report their happiness for granted. Over the course of the history reviewed 

here, substantial efforts have gone into showing that the relevant measures are reliable and 

valid (cf. Angner, forthcoming).  

Finally, the story demonstrates that history can serve to suggest novel experimental and 

theoretical developments. As we saw in Section 2.6, Wilson ended his review by calling for 

more experimental research into the determinants of happiness. While the specific intervention 

he discussed may not be the most suitable one, he was quite right that the possibility of 

exploring experimentally the determinants of happiness is exciting. Moreover, Campbell et al. 

identified interesting avenues of theoretical development when they noted that degrees of 

satisfaction differ radically between individuals even under objectively identical circumstances. 

They also pointed out that this fact is part of the raison d’être of the whole endeavor: ‚If there 

were a close and universal relationship between the level of material possessions and the 

quality of life experience, there would, of course, be little point in undertaking a study of the 

kind in which we are here engaged‛ (1976, p. 10). Campbell et al. suggest that satisfaction may 

be a function not just of actual or perceived attainment, but also of the standard against which 

attainment is assessed. Hence:  

The individual’s assessment may derive from any or all of the following bases of 

evaluation: aspiration levels, or the situation that a person hopes eventually to 

attain, where a given domain is concerned; expectation levels, or the situation he 

feels he is likely to attain in the fairly immediate future; equity levels, or what he 

thinks should be true of his situation if perfect justice prevails, given how much 

he invests in it relative to others; reference group levels, or what he believes to be 

true of the situations of others with whom he identifies, such as friends and 

family or others of his income, race, or occupation; personal needs, or the amount 

of a particular reward he may require, such as how much savings to feel secure, 

how much housing to be comfortable, how much police protection to feel safe; 

and personal values, concerning such intangibles as freedom, equality, and the 

like (Campbell et al., 1976, p. 14).  

While recent research has explored precisely this kind of question, the area is far from 

exhausted.  

It is my hope that this history, though short and in many ways incomplete, will be of 

interest not just to historians of psychology but to practicing social and behavioral scientists, as 

well as to anyone who is interested in the application of the science of happiness to public 

policy. This historical review has offered new perspectives on past and present scientific 

practice and further proof that history can serve to recover knowledge – about natural and 

social phenomena as well as about previous efforts to measure them – that appears to have 

been forgotten; to develop critical awareness of contemporary efforts, their prospects and 

limitations; and to suggest novel experimental and theoretical developments. Further study 

into the history of subjective measures thus seems highly motivated, not just because of what 

we can learn about the historical origins of such measures but also because of what we can 

learn about their philosophical and methodological underpinnings. 

 
Acknowledgements 

This paper is based on Angner (2005). I am grateful to those, too numerous to name, who have offered 

feedback on earlier drafts. All errors remain my own. Unless otherwise noted, all italics in original. 

 

 



The evolution of eupathics 

Angner 

 

www.internationaljournalofwellbeing.org 38 

Author 

Erik Angner 

University of Alabama at Birmingham 

angner@uab.edu 

 

References 

Adams, D. L. (1971). Correlates of satisfaction among the elderly. The Gerontologist, 11(4, Pt. 2), 64-68. 

Allport, G. W. (1935). Attitudes. In C. Murchison (Ed.), A handbook of social psychology (pp. 798-844). 

Worcester, MA: Clark University Press. 

Allport, G. W. (1937). Personality: A psychological interpretation. New York, NY: H. Holt and Co. 

Andrews, F. M. (1989). The evolution of a movement. Journal of Public Policy, 9(4), 401-405. 

http://dx.doi.org/10.1017/S0143814X00008242  

Andrews, F. M., & Robinson, J. P. (1991). Measures of subjective well-being. In J. P. Robinson, P. R. 

Shaver, L. S. Wrightsman, & F. M. Andrews (Eds.), Measures of personality and social psychological 

attitudes, Measures of Social Psychological Attitudes Series (Vol. 1). San Diego, CA: Academic Press. 

Andrews, F. M., & Withey, S. B. (1976). Social indicators of well-being: Americans’ perceptions of life quality. 

New York, NY: Plenum Press. 

Angner, E. (2005). Subjective measures of well-being: A philosophical examination. University of Pittsburgh, 

Pittsburgh, PA. http://etd.library.pitt.edu/ETD/available/etd-08192005-130324/   

Angner, E. (2010). Subjective well-being. Journal of Socio-Economics, 39(3), 361-368.  

http://dx.doi.org/10.1016/j.socec.2009.12.001  

Angner, E. (2011). Are subjective measures of well-being ‚direct‛? Australasian Journal of Philosophy, 89(1), 

115-130. http://dx.doi.org/10.1080/00048400903401665  

Angner, E. (forthcoming). Current trends in welfare measurement. In J. B. Davis & D. W. Hands (Eds.), 

The Elgar companion to recent economic methodology. Northampton, MA: Edward Elgar. 

http://ssrn.com/abstract=1680053   

Argyle, M. (1999). Causes and correlates of happiness. In D. Kahneman, E. Diener, & N. Schwarz (Eds.), 

Well-being: The foundations of hedonic psychology (pp. 353-373). New York, NY: Russell Sage 

Foundation. 

Bateman, B. W. (2001). Make a righteous number: Social surveys, the men and religion forward 

movement, and quantification in American economics. In J. L. Klein & M. S. Morgan (Eds.), The age of 

economic measurement, Annual Supplement to Volume 33, History of Political Economy (pp. 57-85). 

Durham, NC: Duke University Press. 

Bradburn, N. M., & Caplovitz, D. (1965). Reports on happiness: A pilot study of behavior related to mental 

health. Chicago, IL: Aldine. 

Burgess, E. W., & Cottrell, L. S. (1939). Predicting success or failure in marriage. New York, NY: Prentice-

Hall. 

Campbell, A. (1976). Subjective measures of well-being. American Psychologist, 31(2), 117-124. 

http://dx.doi.org/10.1037/0003-066X.31.2.117  

Campbell, A., Converse, P. E., & Rodgers, W. L. (1976). The quality of american life: Perceptions, evaluations, 

and satisfactions. New York, NY: Russell Sage Foundation. 

Cantril, H. (1965). The pattern of human concerns. New Brunswick, NJ: Rutgers University Press. 

Cavan, R. S., Burgess, E. W., Havighurst, R. J., & Goldhamer, H. (1949). Personal adjustment in old age. 

Chicago, IL: Science Research Associates. 

Chang, H. (2004). Inventing temperature: Measurement and scientific progress. Oxford: Oxford University 

Press.  http://dx.doi.org/10.1093/0195171276.001.0001  

Davis, K. B. (1929). Factors in the sex life of twenty-two hundred women. New York, NY: Harper & Brothers. 

Diener, E. (2006). Guidelines for national indicators of subjective well-being and ill-being. Applied 

Research in Quality of Life, 1(2), 151-157.  http://dx.doi.org/10.1007/s11482-006-9007-x  

Diener, E. (2008). Myths in the science of happiness and directions for future research. In M. Eid & R. J. 

Larsen (Eds.), The science of subjective well-being (pp. 493-514). New York, NY: Guilford Press. 

http://dx.doi.org/10.1017/S0143814X00008242
http://etd.library.pitt.edu/ETD/available/etd-08192005-130324/
http://dx.doi.org/10.1016/j.socec.2009.12.001
http://dx.doi.org/10.1080/00048400903401665
http://ssrn.com/abstract=1680053
http://dx.doi.org/10.1037/0003-066X.31.2.117
http://dx.doi.org/10.1093/0195171276.001.0001
http://dx.doi.org/10.1007/s11482-006-9007-x


The evolution of eupathics 

Angner 

 

www.internationaljournalofwellbeing.org 39 

Diener, E., Lucas, R., Schimmack, U., & Helliwell, J. (2009). Well-being for public policy. New York, NY: 

Oxford University Press. http://dx.doi.org/10.1093/acprof:oso/9780195334074.001.0001  

Diener, E., & Seligman, M. E. P. (2004). Beyond money: Toward an economy of well-being. Psychological 

Science in the Public Interest, 5(1), 1-31. http://dx.doi.org/10.1111/j.0963-7214.2004.00501001.x 

Diener, E., Suh, E. M., Lucas, R. E., & Smith, H. L. (1999). Subjective well-being: Three decades of 

progress. Psychological Bulletin, 125(2), 276-302. http://dx.doi.org/10.1037/0033-2909.125.2.276  

Dodge, R. (1930). Autobiography. In C. Murchison (Ed.), A history of psychology in autobiography (Vol. 1, 

pp. 99-121). Worcester, MA: Clark University Press. http://dx.doi.org/10.1037/11401-004  

Easterlin, R. A. (1974). Does economic growth improve the human lot? Some empirical evidence. In P. A. 

David & M. W. Reder (Eds.), Nations and households in economic growth: Essays in honor of Moses 

Abramovitz (pp. 89-125). New York, NY: Academic Press. 

Flügel, J. C. (1925). A quantitative study of feeling and emotion in everyday life. British Journal of 

Psychology, 15, 318-355. 

Frey, B. S., & Stutzer, A. (2000). Maximising happiness? German Economic Review, 1(2), 145-167. 

http://dx.doi.org/10.1111/1468-0475.00009  

Frey, B. S., & Stutzer, A. (2002). Happiness and economics: How the economy and institutions affect well-being. 

Princeton, NJ: Princeton University Press. 

Goldings, H. J. (1954). On the avowal and projection of happiness. Journal of Personality, 23(1), 30-47. 

http://dx.doi.org/10.1111/j.1467-6494.1954.tb02336.x  

Gurin, G., Veroff, J., & Feld, S. (1960). Americans view their mental health: A nationwide interview survey. 

New York, NY: Basic Books. 

Hamilton, G. V. (1929). A research in marriage. New York, NY: A. & C. Boni. 

Hart, H. (1940). Chart for happiness. New York, NY: Macmillan. 

Hartmann, G. W. (1934). Personality traits associated with variations in happiness. The Journal of 

Abnormal and Social Psychology, 29(2), 202-212. http://dx.doi.org/10.1037/h0073108  

Hilgard, E. R. (1957). Lewis Madison Terman: 1877-1956. The American Journal of Psychology, 70(3), 472-

479. 

Kahneman, D., Diener, E., & Schwarz, N. (Eds.). (1999). Well-being: The foundations of hedonic psychology. 

New York, NY: Russell Sage Foundation. 

Kahneman, D., Krueger, A. B., Schkade, D., Schwarz, N., & Stone, A. (2004). Toward national well-being 

accounts. The American Economic Review, 94(2), 429-434. http://dx.doi.org/10.1257/0002828041301713  

Klein, J. L., & Morgan, M. S. (Eds.). (2001). The age of economic measurement. Annual Supplement to 

Volume 33, History of Political Economy. Durham, NC: Duke University Press. 

Larson, R. (1978). Thirty years of research on the subjective well-being of older Americans. Journal of 

Gerontology, 33(1), 109-125.  

Layard, P. R. G. (2005). Happiness: Lessons from a new science. New York, NY: Penguin Press. 

Lebo, D. (1953). Some factors said to make for happiness in old age. Journal of Clinical Psychology, 9(4), 

385-387. http://dx.doi.org/10.1002/1097-4679(195310)9:4%3c385::AID-JCLP2270090418%3e3.0.CO;2-7  

Levy, E. (2001). Quantification, mandated science and judgment. Studies In History and Philosophy of 

Science Part A, 32(4), 723-737. http://dx.doi.org/10.1016/S0039-3681(01)00019-X  

Lyubomirsky, S. (2008). The how of happiness: A scientific approach to getting the life you want. New York, 

NY: Penguin Press. 

McAdams, D. P. (2001). Personality psychology. In N. J. Smelser & P. B. Baltes (Eds.), International 

encyclopedia of the social & behavioral sciences (Vol. 16, pp. 11308-11313). Oxford: Pergamon. 

 http://dx.doi.org/10.1016/B0-08-043076-7/01644-2  

Morgan, A. E. (1934). An attempt to measure happiness. International Journal of Ethics, 44(2), 236-243. 

Morgan, M. S. (2001). Making measurement instruments. In J. L. Klein & M. S. Morgan (Eds.), The age of 

economic measurement, Annual Supplement to Volume 33, History of Political Economy (pp. 235-251). 

Durham, NC: Duke University Press. 

Murray, H. A. (1938). Explorations in personality: A clinical and experimental study of fifty men of college age. 

New York, NY: Oxford University Press. 

http://dx.doi.org/10.1093/acprof:oso/9780195334074.001.0001
http://dx.doi.org/10.1111/j.0963-7214.2004.00501001.x
http://dx.doi.org/10.1037/0033-2909.125.2.276
http://dx.doi.org/10.1037/11401-004
http://dx.doi.org/10.1111/1468-0475.00009
http://dx.doi.org/10.1111/j.1467-6494.1954.tb02336.x
http://dx.doi.org/10.1037/h0073108
http://dx.doi.org/10.1257/0002828041301713
http://dx.doi.org/10.1002/1097-4679(195310)9:4%3c385::AID-JCLP2270090418%3e3.0.CO;2-7
http://dx.doi.org/10.1016/S0039-3681(01)00019-X
http://dx.doi.org/10.1016/B0-08-043076-7/01644-2


The evolution of eupathics 

Angner 

 

www.internationaljournalofwellbeing.org 40 

Myerson, A. (1917). Eupathics: A program for mental hygiene. The Journal of Abnormal Psychology, 12(5), 

343-347. http://dx.doi.org/10.1037/h0073646  

Neugarten, B. L., Havighurst, R. J., & Tobin, S. S. (1961). The measurement of life satisfaction. Journal of 

Gerontology, 16(2), 134 -143.  

Pollak, O. (1948). Social adjustment in old age: A research planning report. New York, NY: Social Science 

Research Council. 

Porter, T. M. (1995). Trust in numbers: The pursuit of objectivity in science and public life. Princeton, NJ: 

Princeton University Press. 

Praag, B. M. S. V. (1968). Individual welfare functions and consumer behavior: A theory of rational irrationality. 

Amsterdam: North-Holland Pub. Co. 

Redelmeier, D. A., & Kahneman, D. (1996). Patients’ memories of painful medical treatments: Real-time 

and retrospective evaluations of two minimally invasive procedures. Pain, 66(1), 3-8. 

http://dx.doi.org/10.1016/0304-3959(96)02994-6  

Rose, A. M. (1955). Factors Associated with the life satisfaction of middle-class, middle-aged persons. 

Marriage and Family Living, 17(1), 15-19. http://dx.doi.org/10.2307/346773  

Sailer, R. C. (1931). Happiness self-estimates of young men. Teachers College, Columbia University 

Contributions to Education, No. 467. New York, NY: Columbia University. 

Seligman, M. E. P. (2002). Authentic happiness: Using the new positive psychology to realize your potential for 

lasting fulfillment. New York, NY: The Free Press. 

Seligman, M. E. P., & Csikszentmihalyi, M. (2000). Positive psychology: An introduction. American 

Psychologist, 55(1), 5-14. http://dx.doi.org/10.1037/0003-066X.55.1.5  

Sen, A. (2008). The economics of happiness and capability. In L. Bruni, F. Comim, & M. Pugno (Eds.), 

Capabilities and happiness (pp. 16-27). Oxford: Oxford University Press. 

Symonds, P. M. (1937). Happiness as related to problems and interests. Journal of Educational Psychology, 

28(4), 290-294. http://dx.doi.org/10.1037/h0061067  

Terman, L. M. (1938). Psychological factors in marital happiness. New York, NY: McGraw-Hill. 

Watson, G. (1930). Happiness among adult students of education. Journal of Educational Psychology, 21(2), 

79-109. http://dx.doi.org/10.1037/h0070539  

Wilson, W. (1967). Correlates of avowed happiness. Psychological Bulletin, 67(4), 294-306. 

http://dx.doi.org/10.1037/h0024431  

Winter, D. G., & Barenbaum, N. B. (1999). History of modern personality psychology theory and 

research. In L. A. Pervin & O. P. John (Eds.), Handbook of personality: Theory and research (2nd ed., pp. 

3-27). New York, NY: Guilford Press. 

Wohlgemuth, A. (1919). Pleasure–unpleasure: An experimental investigation on the feeling-elements. The 

British Journal of Psychology Monograph Supplements (Vol. 2). Cambridge: Cambridge University 

Press.  

 

 

 

 

 

 

 

http://dx.doi.org/10.1037/h0073646
http://dx.doi.org/10.1016/0304-3959(96)02994-6
http://dx.doi.org/10.2307/346773
http://dx.doi.org/10.1037/0003-066X.55.1.5
http://dx.doi.org/10.1037/h0061067
http://dx.doi.org/10.1037/h0070539
http://dx.doi.org/10.1037/h0024431


The evolution of eupathics 

Angner 

 

www.internationaljournalofwellbeing.org 41 

 

T
im

e
li

n
e
 o

f 
e
a
rl

y
 c

o
n

tr
ib

u
to

rs
 t

o
 t

h
e
 l

it
er

a
tu

re
 o

n
 s

u
b

je
ct

iv
e 

m
e
a
su

re
s 

o
f 

w
e
ll

b
e
in

g
 a

n
d

 t
h

e
 s

ci
e
n

ce
 o

f 
h

a
p

p
in

e
ss

. 

A
p

p
e
n

d
ix

 A