Is Happiness Determined By Individual’s Well-Being? Proposing An Econometric Model On Happiness


There is no doubt that each and every individual’s intent is to get to happiness. Most of its decisions are made to accomplish a mental or emotional condition proper to a certain level of well-being. In our opinion, happiness is, in a way, the synonym of well-being, and well-being is a sum of positive situations that matter morally, and of fulfilled needs such as love, power, fun and freedom, stability, general health, education, so on. As many physical illnesses have psychological roots, a good mental condition gets a high importance related to our life and performance. Vitality, enthusiasm and engagement imply dealing with stress, obtaining a positive mental health and a reduced risk of heart disease, therefore a happier and qualitative life. The aim of the paper is to get an answer if either nor is happiness a determinant of our well-being state, if meeting our needs in responsible ways gives us happiness. Therefore, we propose a model in which the left hand side variable “happiness” is determined by a positive thinking and a general state of “feeling good” and “well-being”. The well-being explanatory variables used as determinants in our model include: health, education, employment, age, financial satisfaction, political or religious beliefs, marital status and age.

Keywords: Happinessmental healthsocio-economic factorsutilitywell-being


The word happiness is frequently used and related to different fields of study such as psychology,

biology, anthropology sociology and economics. According to Aristotle, happiness is considered to have

at least two aspects: hedonia (pleasure) and eudaimonia (a life well lived). In modern psychology,

happiness is referred to getting a simple pleasure and meaning, combined with engagement. (Seligman,

2002) Therefore, apart from seeing happiness as “the feeling of being happy,”1 we define it as a

combination of the level of satisfaction with personal life (for example, work, relationships, education,

hobbies) and how good it feels on a daily basis.2 Most psychologists consider happiness as being one of

the six basic emotions3. (Ekman, Friesen, 1971) Others say that happiness has three essential

components: emotion (positive emotions and lack of negative emotions), life satisfaction (purposeful life,

personal growth and well-being) and positive relations with others (love for the others and of the others),

(Arggle, 2001) Also, it is proved that social success and individual’s personality and behaviour are

influencing general happiness. (Saed, Pour Ehsan, 2008) We try to prove below not only the truth of the

affirmations, but which elements are determinants for happiness.

Literature Review

Researches were made to describe the term by correlating it not only with individual’s well-being,

but also with its mental health saying that the progressive development has had consequences for the

human brain. (Angell, 2011) In a 2015 study4, students’ general health status, happiness, self-efficacy,

perceived stress, hopefulness and life satisfaction were measured using self-reported written

questionnaires. The conclusion indicated a significant relation between happiness and psychological well-

being. It was found out that the students having strong relationships and those who enjoybeing with

family and friends are happier and more willing to help the others, theylead their life in a tension–free

manner, they smile and laugh, and sufferof less illnesses, theyenjoy life and acceptthat other peopleare

different and they do not criticize or try to change them, they arecreative, positive, and self-confident. So,

good mental health, life satisfaction and happinessare related. The absence of mental illnessor disorders

caused by high level stress, feeling content and good with one-self, positivism, having choices lead to a

certain level of well-being and finally to happiness.(Glasser,5 1962)

In economics, the approach of happiness is done in terms of utility6. Psychological speaking

utility is correlated with mental health and well-being, thinking that a person that’s useful and active is

more optimistic and positive, therefore, less stressed and mentally healthier,a proper state for being

successful. But utility cannot be observable and cannot be measured. Besides, not all human beings are

1 See Oxford Dictionary as well as other dictionaries.

2 According to A. Parks, in a recent research was suggested that an even-keeled mood is more psychologically healthy than a mood, great heights of happiness achieved regularly are impossible, so the mood goes, as frequently, down.

3 The six basic human emotions are: happiness, surprise, sadness, anger, fear and disgust.

4 The 2015 study was published in the Asian Journal of Psychiatry.

5 William Glasser is psychiatrist and professor, founder of choice theory psychology.

6 The utility theory has been promoted by neoclassical economics as an ordinal notion of measuring human welfare, under the assumption that a rational individual shall ensure its welfare by maximizing the utility.

rational - they are not always maximizing their own satisfaction, but even so they care about their

neighbour, their status in society, and they care about their well-being. There is in people’s nature to

measure their relative welfare, meaning that they measure their own welfare by comparing it with the

others. (Oswald, 1997)

Feelings such as satisfaction or its absence, happiness or unhappiness appear to be of great

importance in individual’s future well-being. Many researchers consider “being happy” as the ultimate

goal of human life. Income, wealth or social status are meaningless if these do not make an individual

happier and healthier, and only together these give a person a recognisable state of well-being. It has

been proved that happiness could be measured subjectively by listening to what human beings have to say

about their state of happiness. A simple survey7 asking people of their level of happiness can give an

honest and real measurement of happiness and subjective utility. So, if happiness is subjective, the stimuli

needed to be happy ” are also subjective, different from one individual to another. Sometimes we have

a misconception about what makes us happy and about the things we could do to increase our level of

happiness and meaning of life, and this comes either from coping the others in terms of needs and

behaviour, or because we don’t know for sure what exactly do we want.

In time, it has been developed a trend in measuring “ utility ” in terms of “ happiness ”8, so

economists have been trying to construct a model to explain the determinants of a person’s happiness, by

using socio-economic variables. Maybe the most important result are those showingthat no matter the

socio-economic and cultural differences, people express similar opinions regarding happiness, but other

results were relevant as well. Reported happiness does not depend on income as much as predicted by

standard utility theory. Although there is a statistically significant positive relationship between income

and happiness - higher incomes and greater happiness are linked (Schnittker, 2008), it was proved that the

relationship is really weak. (Walker, Kavedžija, 2015) However, Easterline (1974) studying the US

population has found that happiness depends on relative income9, not on absolute income of people. Later

on, he got on the conclusion that happiness strongly depends on household health status. Other studies

conducted in Latin America or Europe confirmed Easterline’s findings. (Gerdham, Johannesson, 1997)

Also, Oswald (1997) said that reported happiness is higher among people who are married, healthy,

women, well educated, Caucasians, self-employed, retired and those looking after a home. He observed

that age has an U-shaped impact on happiness: children and elderly are happier than adults, and people

become happier after their 30s. Stevenson and Wolfer (2008) in his studies about race and happiness

stated that there is a gap between black-white happiness, but the gap is decreasing. Other findings

showed that people who have strong religious beliefs or strong political views tend to be happier than

those without these beliefs. Religion persuades happiness for a number of reasons as studies showed:

gives people a sense of purpose and positivism, serves as a resource for dealing with negative life

experiences and existential fears, and not at last, religion promotes a sense of belonging, a social

7 The survey could be simple:”All together, how happy do you think you are: very happy, happy, or not happy”, or more complex: “All together, how happy you would say you are: very happy, quite happy, happy, not very happy, and not at all happy?” 8 Happiness represents a subjective approach for economists in studying human welfare.

9 Relative income depends on income of one individual compared to his friends, neighbours or other people he knows.

connectedness. Thus, statistics studies on social relationships tend to eliminate the association

between religiosity and well-being. Their argument is that religious people report having more social ties,

which being taken into account statistically shows that religion by itself does not predict happiness.

(Seligman, 2002) In 2009, Levinson claimed that happiness relates to not only relative income, but also to

air/ environment quality, some others relate it with individual’s emotional intelligence, his social success

and his personality. (Saed, Pour Ehsan, 2008)

To summarize the works on happiness we are saying that there are common patterns in what

determines happiness. Happiness, and life satisfaction tends to be higher among women, people with

social skills (with diverse activities, with family and/ or lot of friends), very young people and old people,

married and cohabiting people (those who are not alone), the highly educated (finding satisfaction in

their profession), the healthy (there is no dividing between physically and mentally healthy) here being

included people with low blood pressure (who are more calm and less stressed) and people who exercise

especially outdoors (hiking, so on), those with high income (with a certain status and level of well-being),

the self-employed people, those who have sex at least once a week with the same partner, the right-wing

voters, the religious people, members of non-church organizations, volunteers, and those who live in

western countries. (Blanchflower, 2008)

Developing a Model On Happiness


In order to see the relevance of a certain variable, if it determines or not happiness we propose to

develop a model - a difficult task in econometrics. A statistically significant variable may have no

economic reason to be included in the model and vice versa. (Gujarati, 1995) So, based on the literature

we identify few representative variables for making people happy or unhappy. Most of these variables are

proving a certain level of well-being. We include them in a table alongside the possible answers: level of

education, health status, relative financial position/income, marital status, sexual life, gender, race, age,

work status, air and environmental quality, vacation, religious devotion and political view. (Table 1 )

Table 1 -
See Full Size >

The first step in the model development is to find a statistical community and to apply the

questionnaires to the selected people. Data should be manipulated which means that we delete the

observations with missing variables, keeping only the complete, qualitative data that would be used and

checked for meaningfulness.

Second step includes choosing the model. We think of using a logistic regression since the

response variable is binary10, on the model:

z = β 0 + β 1 x 1 + β 2 x 2 + β a x a + β 4 x 4 + + β n x n (1)

This model is fitted to the equation:

f z = 1 1 + e - z (2)

STATA is used to look if our variables

are positively correlated with happiness and statistically significant as we were expecting.

A brief description of the variables shows the following:

Happiness : The dependent variable happy is a qualitative variable, therefore it is used to conduct the analysis;We create the variable d_happy that takes the value 1 if the respondent claims to be very or pretty happy, and the value 0 if the answer was not happy. (Table 2 )

Explanatory variables : All the right hand side variables are transformed into binary variables to conduct the logit and probit estimations . We choose relevant variables we are being confident about.

Our proposed model becomes: d_happy=βo+β1d_health+β2d_race+β3d_nat+β4d_age+β5d_educ+β6d_sex+β7d_m+β8d_satfin+β9d_wrkst+β10d_pers+β11d_gender+β12d_park+β13d_vac+β14d_attend+β15d_polview+error

10 A linear regression would give doubtful results since conditions like homoscedasticity are not being satisfied.

We have:

Table 2 -
See Full Size >

A logistic regression is being run focusing on certain variables that we think they can predict



The results from STATA show the convergence, if our variables are positively correlated with

happiness and statistically significant as we were expecting, if they are explanatory variables on happiness

or not. The interpretation should be rather straightforward indicating the percentage of confidence interval

on the positive real line which shows the variable as being indeed an indicator of happiness.


A model is judged based on how well its right hand side variables can predict the variable happiness

using the estat classification command in STATA with several cut-off values that will maximize the

specific number. We also run a logistic regression on doubtful variables, seeing whether or not these

variables can predict happiness with a percentage of confidence level, if they are statistically significant

and positively correlated. Next, we run a third logistic regression with the merged significant variables.

The results could show us that in the merged model some variables are no longer statistically significant.

However, a high percentage of correctly predicted response variables could show an improvement from

the previous attempts. All other variables remained statistically significant.

Finally, to make sure that the found significant variables have a statistical impact on happiness, it is

useful to re-run the logistic regression with these variables and to analyze the results.

Diagnostic testing is run for heteroscedasticity,11multicollinearity,12 and the goodness of fit. The

magnitude of significant variables allows us to understand the degree of importance of each explanatory

variables and direction of relationship, and to interpret the marginal effect in terms of probability.


The model we ended up with is considered by us simple and good, but only after running it we

have the level of prediction, the percentage of correct classification, the number of explanatory variables

and an actual conclusion about the correlation between happiness the explanatory variables. Then we

could answer to questions such as: is happiness determined by individual’s well-being, which individual

is more likely to be happy: the richer, the healthier, the most faithful one? Until we actually run the

model on real data to see its relevance we assume that there is a certain correlation between happiness and

well-being represented by health, as a general physical health, age, social success and relative financial

situation and work status, and a relative correlation of happiness with race, nationality, love life, level of

education or political and religious beliefs.

11 If cross sectional data is used, there is a possibility of heteroscedasticity in the model. Heteroscedasticity means variance of error term is not constant across observations. If heteroscedasticity is found in this model, then should be used robust heteroscedasticity standard errors technique s.

12 It is calculated the Variance Inflation Factor (VIF); if a problem is detected some explanatory variables can be dropped.


  1. Angell M. (2011). The Epidemic of Mental Illness: Why? The New York Review of Books,58, 11
  2. Arggle M. (2001). The psychology of happiness. (Rev.ed.) Routledge. East Sussex. Britain.
  3. Blanchflower, & D.G., Oswald, A.J. (2008). Is well-being U-shaped over the life cycle? Social Science &Medicine, 66, 1733-1749.
  4. Easterline, R.A. (1995). Will raising the incomes of all increase the happiness of all? Journal of Economic Behavior & Organization, 27, 35-47.
  5. Ekman, P. & Friesen, W.V. (1971). Constants across cultures in the face and emotion." Journal of Personality and Social Psychology. 17, 124–129.
  6. Gerdham,U.G., & Johannesson, M. (1997). The relationship between happiness, health and socio-economic factors; results based on Swedish micro data. Working Paper series in Economics and Finance.
  7. Glasser,W. (1962). Mental Health or Mental Illness? Psychiatry for Practical Action
  8. Gujarati D.N. (1995). Basic Econometrics. McGraw-Hill.
  9. Levinson, A. (2009). Valuing Public Goods Using Happiness Data: The Case of Air Quality. NBER Working Paper 15156.
  10. Oswald, A.J. (1997). Happiness and Economic Performance. The Warwick Economics Research Paper Series
  11. Saed, O., & Pour Ehsan, S. (2008). Relational studying of components of emotional intelligence or psychological health on students of Teacher Training University in Tehran. Mental Health Seminar presentation.
  12. Seligman, M.E.P. (2002). Authentic Happiness: Using the New Positive Psychology to Realize Your Potential for Lasting Fulfillment. New York Free Press.
  13. Schnittker J. (2008). Happiness and Success: Genes, Families, and the Correlates of Well-Being. American Journal of Sociology, 114, S233-S259
  14. Stevenson, B. & Wolfer, J. (2008). Happiness Inequality in the United States. NBER Working paper 14220.
  15. Walker, H. & Kavedžija, I. (2015) Values of happiness. Journal of Ethnographic Theory, 5, 8

Copyright information

This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (, which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

About this article

Cite this paper as:

Click here to view the available options for cite this article.


Future Academy

First Online




Online ISSN