(1) Background
Research into the relationship between money and happiness has yielded mixed results (Killingsworth et al., 2022, p. 1; Ford et al., 2016). One reason for these discrepancies may be differences in study designs, such as sample characteristics and how income and well-being are measured. In fact, at least one replication study has demonstrated that the relationship between money and happiness depends on methodology used in the study and, when the relationship exists, it may be weak (Aknin et al., 2020). Further, many studies have focused solely on the relationship between income and positive affect, though some researchers suggest that a more complete answer may lie in analyzing negative affect as well (Killingsworth et al., 2022).
Diener and colleagues (2010) aimed to provide a broader understanding of the relationship between income and well-being by incorporating multiple indicators of happiness. Rather than relying on a single measure, the study included life evaluations, positive and negative emotions, income, and material possessions. This multi-faceted approach allows for a more nuanced exploration of how money and happiness may be connected (Disabato et al., 2016, p. 475). Prior research has emphasized that the relationship between income and psychological well-being is not unidirectional and may be influenced by unmet psychological needs and access to mental health resources (Srivastava et al., 2001, p. 961; O’Donnell et al., 2021). Diener and colleagues (2010) included measures of both emotional states and life satisfaction, offering a unique opportunity to further explore these bidirectional dynamics.
The data in this dataset come from a multi-lab collaboration supported by the Collaborative Replication and Education Project (Project CREP; Wagge, 2019) designed to replicate the original Diener et al (2010) study. This dataset is a combination of data from several sites and were collected primarily by teams of undergraduate researchers. The dataset includes variables related to basic and psychological needs and individual wellbeing. Additionally, these data include demographic information collected from participants at different study sites and income adjusted to United States Dollar values from October 2023 to account for any differences in currency across multiple currencies and multiple time points.
Data were collected by the following Institutions all over the world: University of Vienna – Austria, Illinois Institute of Technology in Chicago Illinois, Marian University of Indianapolis, University of Toronto, Education University of Hong Kong, Tilburg University – Netherlands, and Pacific Lutheran University (as seen in Table 2). Materials were provided to research teams via an OSF page (https://osf.io/px0sw/). Participating research teams forked OSF project pages and stored all materials and data on these pages. Following CREP procedures, projects were reviewed prior to data collection. Some projects included additional measures (a “direct+” replication); teams describe these additional measures on their project wikis. Any additional measures were required to be added at the end of the survey.
The data presented in this paper can be used for both research and educational purposes. Researchers may find this dataset helpful to conduct exploratory analyses on predictors of life satisfaction and educators may find this dataset useful for in-class examples in statistics or methods courses. For instance, the data could be used in statistics courses when learning about correlation and regression. That way, students can form hypotheses about the relationship between happiness and wealth and then test the relationship with the data. Educators may also use these data in social psychology or wellbeing courses, as they can be used to demonstrate how researchers study these topics.
(2) Methods
2.1 Study Design
This dataset was created by combining data from all replications of Diener et al (2010) through Project CREP. Overall, 58 different research teams created forks of the Open Science Framework Page (https://osf.io/px0sw/). Twenty-one of these replications were publicly accessible. To ensure we had as much data as possible, we reached out to research groups without publicly accessible data, though most of these contact attempts were unsuccessful. Of the 21 publicly available replications, 11 were excluded for the following reasons: ten did not provide any data or summary statistics and one was missing a codebook, which made the data uninterpretable. Descriptions of the included replication study sites are available in Table 1. One study site collected data via phone calls, and the remaining sites collected data online.
Table 1
Replication Study Sites and sample sizes.
| STUDY SITES | DATE COLLECTED | SURVEY MODALITY | N |
|---|---|---|---|
| (1) Replication of Diener et al. (2010) at University of Toronto, tut5401, W19 (Tian et al., 2024) | Apr. 2019 | Online | 57 |
| (2) Wealth and Happiness: Replication of Diener, E., Ng, W., Harter, J., & Arora, R. (2010) at University of Vienna (Müller et al., 2021) | Oct. 2020 | Online | 163 |
| (3) Replication of Diener, Ng, Harter, & Arora (2010) at IIT Spring 2018 (Pys, P. M et al., 2021) | Jan. 2018 | Online | 213 |
| (4) Wealth and Happiness: Replication of Diener, et al. at IIT 2019 (Legate et al., 2022) | Mar. 2019 | Online | 183 |
| (5) Replication of Diener et al. (2010) at Marian University (Reed et al., 2020) | Jan. 2020 | Online | 127 |
| (6) Replication of Diener et al. (2010) at Education University of Hong Kong (Wong et al., 2019) | Dec. 2019 | Phone | 200 |
| (7) (Pierce County) Replication of Diener E., Ng W., Harter J., & Arora R. (2010). for Collaborative Replications and Education Project (Hatton et al., 2019) | Jan. 2014 | Phone | 100 |
| (8) CREP Diener Replication Project. Tutorial Section TUT5401. W19 (Merkley et al., 2019) | Apr. 2019 | Online | 29 |
| (9) Replication of Diener, et al. (2010). at Pacific Lutheran University (Grahe et al., 2020) | Apr. 2019 | Online | 99 |
| Total N | 1171 |
[i] Note. The table includes information about all replication sites, survey modality and sample size. An OSF link is included for each specific page.
All data collection sites followed a similar procedure. Sites around the world volunteered to directly replicate Diener et al (2010) using the same materials available on the CREP OSF page (https://osf.io/px0sw/overview). Specifically, all participants completed an income measure, and surveys related to their basic and psychological needs. Some sites chose to include extension hypotheses and included additional measures to test these hypotheses following the original study. Prior to data collection, the site’s procedure was reviewed by two CREP reviewers and an executive team member to ensure that the direct replication was occurring consistently across sites.
2.2 Time of data collection
The replications occurred between 2014 and 2018 (see Table 1). We aggregated the data across sites in October 2023.
2.3 Location of data collection
The replications included in this dataset were collected from several different Universities throughout the United States, Canada, Austria, Netherlands, and China. The four replications in the United States were from Marian University in Indianapolis, Indiana, Pacific Lutheran University in Tacoma, Washington, and two of the four replications were from the Illinois Institute of Technology in Chicago, Illinois. There were two replications from the University of Toronto, Canada. The rest of the replications were from the Education University of Hong Kong, China, the University of Vienna, Austria, and Tilburg University, The Netherlands. The aggregated dataset was combined at the University of Minnesota – Twin Cities in October 2023.
2.4 Sampling, sample, and data collection
The overall sample contained 1,171 participants. Demographic information for the sample is reported in Table 2. While most study sites reported demographic information, several study sites opted out of reporting some demographic data. Study sites 4 and 6 did not report age in their datasets. Study sites 4, 5, and 6 did not report the marital status or residential area of their participants and study sites 6 and 7 did not report education attained. Lastly, study site 9 did not report marital status using the key from the Diener et al. (2010) paper, thus it has been omitted from the martial status table. We could not contact the authors, and marital status was not included in their analyses on OSF.
Table 2
Demographics of total study sites.
| CHARACTERISTICS | CATEGORY | FREQUENCY | PERCENTAGE |
|---|---|---|---|
| Gender | Man | 617 | 52.7 |
| Woman | 512 | 43.7 | |
| Nonbinary | 4 | 0.3 | |
| No response | 38 | 3.2 | |
| Marital Status | Divorced/Separated | 2 | 0.17 |
| Domestic Partnership | 144 | 12.3 | |
| Married | 23 | 1.9 | |
| Single | 431 | 36.8 | |
| NA | 472 | 40.3 | |
| Residential Area | Rural | 34 | 2.9 |
| Suburban | 172 | 14.6 | |
| Urban/City | 485 | 41.4 | |
| Village | 20 | 1.7 | |
| NA | 460 | 39.3 | |
| Education | Graduate | 41 | 3.5 |
| Full Tertiary | 208 | 17.8 | |
| Some Tertiary | 461 | 39.4 | |
| Secondary | 163 | 13.9 | |
| Primary or less | 31 | 2.6 | |
| NA | 259 | 22.1 | |
| Other | 8 | 0.68 |
[i] Note. The table includes information about all replication sites, gender, marital status, residential area, and education.
Subsequently, 958 participants reported income, 81.8% of the total sample. The proportion of participants reporting income varies across study sites (see Table 3). Median annual income (USD), calculated among those reporting income, ranged from $3,650 ($0 – $43,800) to $80,000 ($40,000 to $125,000). Currency across study sites were adjusted to the equivalent United States dollar value in October 2023.
Table 3
Descriptive statistics for total income and income per study site.
| STUDY SITE | N | INCOME REPORTED n(%) | MEDIAN INCOME (USD) | IQR |
|---|---|---|---|---|
| Total | 1171 | 958 (81.8%) | 40,000 | 13,000–85,750 |
| Site 1 | 57 | 57 (100%) | 3,650 | 0–43,800 |
| Site 2 | 163 | 139 (85.3%) | 16,350 | 10,900–38,150 |
| Site 3 | 213 | 150 (70.4%) | 70,000 | 33,000–108,750 |
| Site 4 | 183 | 137 (74.9%) | 75,000 | 45,000–160,000 |
| Site 5 | 127 | 79 (62.2%) | 50,000 | 19,150–100,000 |
| Site 6 | 200 | 193 (96.5%) | 23,400 | 0–52,000 |
| Site 7 | 100 | 97 (97%) | 36,000 | 65–60,000 |
| Site 8 | 29 | 29 (100%) | 17,520 | 6,570–138,700 |
| Site 9 | 99 | 77 (77.8%) | 80,000 | 40,000–125,000 |
[i] Note. Table 3 includes income data from the combined study sites and each individual site. The total N is reported for each site, as well as the total income reported n (%), median income (USD) and interquartile range (IQR).
2.5 Materials/survey instruments
The replications we included in the aggregated dataset used the same materials and surveys as the original Diener and colleagues (2010) study. More details about the surveys are available on our OSF page (https://osf.io/qdx7p).
Cantril Scale
The Cantril Scale was used to examine an individual’s perception of their well-being (Cantril, 1965). This measure quantifies life satisfaction by having participants imagine a ladder with 10 rungs. The top rung (10) represents the best possible life quality, and the bottom rung (1) represents the worst possible life quality. Participants are instructed to pick which rung of the ladder they feel best indicates their current level of life satisfaction. The Cantril Scale has strong convergent validity with other well-being measures across diverse samples.
Day Construction Method
The Day Construction Method (DCM), which evaluates how often a person spends their time doing different activities on a day-to-day basis, was used to examine how people experience their lives daily (Daniel et al., 2004, p. 1776) and is generally considered a valid survey tool in assessing well-being and daily experiences. The first DCM Survey given to participants asked six emotion related questions. Participants were instructed to respond “yes” or “no” if they remembered engaging in one of the emotions or behaviors during the previous day. For instance, participants were asked questions like, “Did you smile or laugh a lot yesterday?” and “Did you experience depression during a lot of the day yesterday?”.
Participants were again asked a different set of questions following the Day Construction Method format. These were five questions relating to activities and relationships that the participant engaged in the previous day. For example, “Did you learn something new yesterday?” and “Do you have family or friends you can count on in an emergency?” Both surveys aim to seek the participants’ emotional well-being, and support system.
Household Income Survey
This study also included questions related to household economic status. Participants were asked a single question about their households’ income and the number of people in the household. Participants also responded to three questions about their ability to access television, computer, and internet (yes = 1, no = 0). These responses were averaged together to create a single score reflecting their economic status. Participants answered two questions about access to basic needs, specifically whether they had access to food or shelter. These responses were averaged (yes = 1, no = 0). Finally, participants were asked if they were satisfied with their standard of living.
Assessing Short Term Financial Confidence:
In addition to the well-being measures from Diener et al. (2010), the replication protocol included two social status items (e.g., the ability to pay a $300 speeding ticket or $5000 lawsuit within a week). These questions, created as part of a replication and extension project conducted by CREP (see https://osf.io/ar5fc/), were added to the end of the questionnaire to gauge the financial stability of the participants and were calculated into their overall financial well-being.
2.6 Quality Control
We began reviewing various OSF forks linked to the project OSF page (https://osf.io/px0sw/). Many of these forks were inaccessible or set to private. Out of the 21 publicly available replications, we excluded 12 due to the absence of raw data, unresponsive pages, missing summary, descriptive information, or results. One additional dataset was excluded because the raw data were not properly labeled, making it impossible to determine key variables. Data cleaning was conducted using excel and RStudio, the cleaned. csv file is available on our OSF page along with the corresponding R Markdown file used for preparing the data for analysis.
2.7 Data anonymization and ethical issues
All data were accessed through OSF and open for public use with no identifying information in which all individual sites received IRB approval for their methods. Ethics approval or a waiver of required ethics approval was available for each site via their specific OSF page listed in Table 1.
2.8 Existing use of data
The dataset was used to conduct a pooled analysis that was presented as a poster at CREPCON 2024.1 To our knowledge, these data have not been published in any other way.
(3) Dataset description and access
3.1 Repository Location
OSF General Page: https://osf.io/qdx7p/overview.
Github repository: https://github.com/odcosta20/Pooled-Data-Diener-et-al-2010.git.
3.2 Object/File Name
Diener_et_al_replication_data.csv.
3.3 Data type
The data is secondary data compiled in OSF. The data contains predominantly categorical variables with income being a numerical variable. The data contains psychological, self-report, and socio-economic measures that are all self-reported.
3.4 Format names and versions
The version of R used is, R version 4.5.2 (2025-10-31). Materials for this dataset are found in the OSF link above. Specifically, the Data folder inside the Methods and Materials folder. Within the Data folder, there is a PDF of the Codebook, an HTML file called Diener et al replication data cleaning, an .csv file called Diener_et_al_replication_data, and a README.rmd as well. We have also inclueded a “cleaned_diner_et_al_dataset.csv”, this is the dataset that has been cleaned after it has ran through the Rscript. Packages used are readxl, dplyr, tidyverse, ltm, ggplot2, tidyr, gridExtra, knitr, markdown, and kableExtra.
3.5 Language
English
3.6 License
CC-By Attribution 4.0 International
3.7 Limits to sharing
N/A
3.8 Publication date
The data was made public on OSF on 12 November 2023.
3.9 FAIR data/Codebook
To adhere to the FAIR guidelines—Findable, Accessible, Interoperable, and Reusable— when utilizing this data, we have undertaken the following actions.
Findability
To ensure the data and necessary materials are findable, a public repository has been created in Open Science Framework for this project titled Compilation of Diener Et Al. (2010) OSF Multi-Site Replication Projects. It can be accessed directly at https://osf.io/qdx7p/ as well as assigned a DOI: 10.17605/OSF.IO/QDX7P. The repository includes appropriate metadata, keywords, and subject categories to aid with indexing and visibility through repository search capability. Additionally, there is a Github repository (https://github.com/odcosta20/Pooled-Data-Diener-et-al-2010.git) containing the Codebook, the .csv file, the .html file, and the README.rmd file.
Accessibility
The data are openly available under an open-access license, Creative Commons Attribution 4.0 International. This allows unlimited access and use, if proper attribution has been given to the authors. The data can be downloaded using standard internet browsers without the need for special authentication. There should be minimal restrictions as all data is anonymized and does not contain personally identifiable information. The institutions that collected the original data are cited properly within the repository.
Interoperability
The data is stored in a generic .csv file to allow for compatibility across a variety of tools and software platforms. The variables were labeled using simple and descriptive names that inherently reflect the subject matter. No formal or structured ontologies were explicitly used; however terms were used to be consistent with common usage within this field of work. Additionally, the metadata is given in a README file and a codebook which allows clear interpretations and potential use of the data for further work.
Reusability
The data is under a clear open license (Creative Commons Attribution 4.0 International, CC BY 4.0), allowing others to reuse the current data. A detailed codebook has been attached in the repository as a pdf titled Diener et al replication data codebook. The codebook includes variable names, descriptions, value labels, data type, and category names for categorical variables. The dataset follows community standards related to the field, ensuring compatibility and ultimately reusability.
(4) Reuse potential
These data could have several potential uses. First, researchers may explore relationships between variables using this dataset. With contribution from multiple international sites, the dataset offers a large sample with high variability, and generalizability across populations. It is also openly available to the public on our OSF page (https://osf.io/qdx7p/), enabling replication and secondary analyses. However, it is important to note that income data were not collected from all participants in several replications, which may impact certain analysis and aspects of data quality. Additionally, many of the participants were college-aged students regardless of data site, this would explain the lack of income data as many young adults might not have an income and participated in the studies.
These data can also be used for teaching purposes. They are well suited from use in introductory statistics courses to demonstrate correlation and regression analysis and may be valuable in social psychology classes exploring the relationship between types of needs and satisfaction.
Notes
Acknowledgements
Finally, we thank Project CREP for selecting this study to be replicated, providing the infrastructure for replications to occur, and reviewing the materials of all replication projects.
Author Contributions
The following individuals made valuable contributions to the development, collection, and processing of this dataset, and we gratefully acknowledge their efforts. Olivia Costa contributed to data collection/cleaning, analysis, manuscript drafting and editing, creating the project OSF page and GitHub repository page, and poster for presentation. Quincey Feragen contributed to data collection, analysis, manuscript editing, and poster presentation. Penelope Corbett and Theodore Rodgers contributed to the data analysis, cleaning, and poster presentation. Dr. Amanda M. Woodward contributed to the design and organization of the course through which this project was developed, providing ongoing guidance throughout the research process, and editing of this manuscript. Dr. Jordan Wagge provided guidance on this project and edits to this data paper. These study sites and their contributors’ gathered data so that we could use it for our analysis. The Vienna study site contributors: Martin Müller, Andre Julian Hartmann, Malte Petersen, Johannes Ayrle, Johanna Panse, Miriam Pichler, Markus Besser, Stefan Keinrath, Gregor Armbruster, Alexander Pfemeter, Regina Gruber, Matthias Winkler, Julia Rager, Laura Schock, Veronika Johanna Ellensohn Martin Meisterhofer, Florian Siepmann, Lucia Pfeifer Kristina Huemer, Melanie Maurer, Nicolas Pils, Stefan Schwaiger, Charlotte Eppenberger, Anna Viktoria Ratheiser, Stefanie Krauth, Pascal Keller, Christian Becker, Thomas Gebetsberge, and Angela Filz. The Illinois Institute of Technology in 2018 study site contributors: Medha, Paulina Pys, Brittany Young, Gabriel Corzo, Nicole Legate, Frank Shu, and Dheeksha Ranginani. The Illinois Institute of Technology in 2019 study site contributors: Nicole Legate, Kristi Johnson, Ashley Aguilar, Jessica Park, Gitika Chalasani, Dheeksha Ranginani, Sobia Sultana, Kaushik Suryanarayanan, and Magda Wisniewska. The Marian University study site contributors: Zachary Reed, Payton Haygood, Taylor Petit, Bryce Lang, and Amanda Egan. The University of Toronto study site contributors: Zhiyue Tian, Molly Metz, Yihan Qiu, Luanzhenxi Yang, Zhousiyu Li, and Luman Zhong. The Education University of Hong Kong study site contributors: Caroline Wong and Wilbert Law. The Tilburg University study site contributors: Cody James Hatton, Jon Grahe, and Hale Gervais. The Pacific Lutheran University study site contributors: Jon Grahe, Heidi McLaughlin, Megan Psick, Rudy King, Ricky Haneda, HeiTung Fung, Zayde Vetter, Devin Johns, and Farhang Hesami. The University of Toronto study site contributors: Jillian Merkley, Mathura Kugan, Isabella Adamo, and Hugh Yoon.
Peer Review Comments
Journal of Open Psychology Data has blind peer review, which is unblinded upon article acceptance. The editorial history of this article can be downloaded here: