World Database of Happiness, Correlational Findings


Abstracts part of this page, full text as PDF files in new window.

Full contents Contents
Chapter 1

Plan of this collection

Full text
Chapter 2

Selection of studies

Full text
Chapter 3

Notation of findings

Full text
Chapter 4

Statistics used

Full text
Chapter 5

Classification of findings

Full text
Chapter 6

Uses of this collection

Full text

Text by Ruut Veenhoven and Wim Kalmijn (chapter 4), September 2002
Revised version of  'Correlates of Happiness', volume 1, 'Plan of the data-collection', RISBO, Rotterdam Netherlands, 1994, ISBN 90-72597-47-8.

Chapter 1

This catalog is part of the ‘World Database of Happiness’, which contains results of empirical investigations on 'happiness', in the sense of 'life-satisfaction'.

This catalog presents correlational findings, that is, observations about conditions that differ systematically between happy and unhappy persons. These findings on happiness are summarized in standard abstracts, which provide information on measurement, statistics and sampling. These mini-abstracts are presented in this database.

Abstracts are rubricated in subject-categories. Within each subject-category, the findings are ordered by nation; within nations the findings are ordered by year of investigation. This is to facilitate comparison across time and culture. This catalog presents the subject-categories in alphabetical order. Browsing the subject-index allows an overview.

The aim is to bring together the scattered empirical findings on happiness to prepare for synthetic analysis. This endeavor differs from earlier synthetic studies in that it is more homogenous and complete


Chapter 2

Happiness is defined as the degree to which an individual judges the overall quality of his life-as-a-whole favorably. Within this concept two 'components' of happiness are distinguished: hedonic level of affect (the degree to which pleasant affect dominates) and contentment (perceived realization of wants). These components represent respectively 'affective' and 'cognitive' appraisals of life and are seen to figure as subtotals in the overall evaluation of life, called overall happiness.

Happiness as defined here can be measured by means of questioning and hedonic level also by observations of non-verbal behavior. Though happiness is measurable in principle, not all the questionnaires and observation schedules used for its measurement are deemed acceptable. Many measures tap in fact broader pheno­mena than defined here. These measures are left out in this database of happiness. All the findings reported here are based on queries that successfully passed a test for face-validity.


Chapter 3

Standard-excerpts of each research-report are made. These excerpts consist of three parts, which can be retrieved independently.  

Part 1  Study ·

This part reports never more than one study  

Part 2  Measured happiness

If the study contains more than one indicator of happiness, this part reports these separately.  

Part 3  Correlational findings (finding-abstracts)

If a study relates happiness to more than one variable, this part presents more than one mini-abstract.

This chapter describes what information is comprised in each of these parts and defines the technical term used. 


Chapter 4

This chapter starts with a short review of statistical analysis and then presents an overview of  statistics used in the excerpts. These statistics are presented in the two schemes below. All are described in an appendix.  

Overview of statistics and procedures for bi-variate situations  
Level of measurement of the correlate                  Level of measurement of the happiness response
            Ordinal             Metric         Dichotomous
Metric  (§ 4/3.1.3.) Correlation coeff.  (r)
Correlation ratio (R²)
Regression coeff. (b)
Beta coefficient (b)
Ordinal  (§ 4/3.1.2.) Spearman’s rank   corr. coeff.(rs)
Kendall's tau-a (ta)
Kendall's tau-b (tb)
Kendall's tau-c (tc)
Goodman/Kruskal's tau Gamma (G)
Somers’ D (Dyx)
Nominal (§ 4/3.1.1) Chi-square (c²)
Pearson’s C
Cramér's V
Tschuprow's T
One-Way Analysis of  Variance withmultiple comparisons(BMC, DMRT, SNK) Correlation ratio (E²)
Dichotomous (§4/3.1.1) Difference in % (D%)
Difference modus (DMo)
Difference means (DM)
- transformed (DMt)
- standardized (DMs)
Critical ratio  (CR)
Cohen's d
Hedges's g
Point biserial correlation (rpb).
Correlation ratio (E²)
Fisher's 2x2 test
Odds ratio  (OR)
Yule's Q
Yule's Y
Logit coefficient (lgt)
Gamma (G)


Overview of statistics and procedures for multivariate situations

Statistical procedure                       Statistics
Two or more  correlates, all-metric
Single happiness
Multiple Regression Analysis Multiple correlation coefficient  (R)
Coefficient of determination (R²)
Adjusted coefficient of determination (Ra²)
Partial regression coefficient  (b)
Standardized partial regression  coeff. (b)
Partial correlation coefficients (rpc)
At least one correlate metric and at least onenominal
Single happiness
Analysis of Covariance (ANCOVA) Difference in adjusted means (Ma)
Coefficient of determination (R²)
Regression coefficient (b)
Standardized regression coefficient (b)
Two or more nominal correlatesSingle happinessResponse One-Way Analysisof Variance (ANOVA) F-test and multiple comparisons.
Loglinear models
Contingency Tables Chi-square
Two or more happiness responses. Multivariate Analysis ofVariance (MANOVA) Hotelling's T2


Chapter 5

This catalog stores standard abstracts of correlational research findings. These abstracts are categorized in several ways and can thereby be easily retrieved. Findings are classified for:

  1. Subject matter, that is, the variable related to happiness
  2. Measurement of happiness
  3. The kind of people investigated
  4. Methodological characteristics of the investigation.

These classifications are presented in this chapter


Chapter 6

Information about correlates of happiness is of course relevant for a better understanding of happiness and provides information for policies aiming at greater happiness for a greater number. The data in this catalog can contribute to a better understanding of other matters as well.

Methodologically, the data-collection can be used in three ways: for the integration of available research, for theory development and for orientation on new research. All these applications make research-effort more cumulative.

This data collection is meant for the scientific community in the first place. Further it is also of interest for policy makers and the general public

Like any tool, this data-catalog has its pros and cons. Its qualities should be compared to alternative sources of information about research-effort in the field; that is, with narrative literature-reviews on happiness, and with data banks that allow secondary analysis of surveys that involved indicators of happiness