Which do you think would make you happier?
Spending time with a close friend
Spending the same amount of time with three close friends (separately or together)
Spending the same amount of time with a close friend, a family member, and a work colleague (separately or together)
Most of us have heard that spending time with people is correlated with well-being. But this recent paper found that well-being is also correlated with the variety of kinds of relationships you spend time with. In other words, you could be better off getting a mix of time with friends, family, colleagues, etc., rather than spending all your time with one category.
As part of our effort to combat the replication crisis in psychology via our Transparent Replications project, we ran a replication study to check this result.
It turns out that, in spite of some problems we found in the original study with how they calculated their results, when we redid their study and corrected the mistakes we replicated the finding! This suggests that the diversity of kinds of social interactions someone has is a predictor of their well-being (at least, for people in the U.S. where we performed the replication).
In this email, we’re going to break down our replication study, as well as our evaluation of the original, but if you just want to know how to use these results to be happier, you can skip to the section "What this means for you".
Summary of Study and Results
The original study had 578 participants, and our replication study had 961. Each participant:
Answered two questions about their subjective well-being (i.e., happiness) over the last 24 hours.
Filled in a diary for their previous day, reporting between 3 and 9 episodes of activity.
For each reported episode, participants indicated whether they were alone, socializing in person, or socializing virtually.
For each episode of in-person socializing, participants were asked to report which relationship types they had with the other people present (a significant other, adult child, young child, friend, etc.)
Two numbers were then computed for each participant:
Social portfolio diversity: a measure of the variety of relationship types a person interacted with and how evenly their interactions were distributed among them.
Social interaction: the total amount of in-person interaction time the participant had in the day. This was used as a "control" to help make sure that the effects of spending more time with people in general could be distinguished from the effects of a diversity of relationship types.
The dataset was then analyzed using linear regression and the results replicated the results of the original study: we found a statistically significant positive relationship between social portfolio diversity and well-being (β = 0.095, b = 0.410, 95% CI [0.085, 0.735], P = 0.014, n = 961). This is what the original study claimed and, hence, we consider the original finding to have replicated.
This appears to be evidence that having a variety of kinds of social interaction has additional benefit, over and above the benefit of the amount of social interaction you have.
Our verdict on the original study
We rate studies according to three criteria: transparency, replicability, and clarity. Here’s the breakdown of how we rated the study in the original paper.
Transparency
✅ The authors of this study provided their data and experimental materials through the Open Science Framework, which earns this study points in transparency.
❌ Another point in the study’s favor is that it was pre-registered. Unfortunately, the pre-registered hypothesis did not include their control variable.
❌ Also, the study’s analysis was not dealt with transparently: no analysis code was provided, and the authors did not provide a clear response to some inquiries about a key analysis question that left unanswered in the paper and supplemental materials.
Replicability
✅ We were able to replicate this study’s finding, despite some significant issues with clarity (see below). We awarded this study 5 stars for replication because its key finding met the criteria for replication that we outlined in our pre-registration.
Clarity
✅ The analysis used in this study is reasonable and not unnecessarily complex.
❌ The computation of social portfolio diversity is described in three inconsistent ways: one way in the study’s pre-registration, another way in the paper’s introduction, and a third way in the paper’s Materials and Methods section. By carefully looking through the data, we were able to figure out which of these methods of computation was actually used and replicate it. However, this computation also seems to be incorrect in an important respect.
❌ The sample size reported in the paper for the study is incorrect due to excluded cases based on the miscalculation of social portfolio diversity.
❌ The calculation of the control variable is not conducted the way it is described in the paper, and appears to be miscalculated.
❌ The study does not report an R2 value for its regression analyses, which obscures the small amount of the variance in the dependent variable that is explained by their overall model and by their independent variable specifically.
What this means for you
The main finding of the original study did replicate, providing some evidence for the claim that increasing the diversity of your social interactions could improve your well-being.
Of course, we unfortunately can't be sure that interacting with more different types of people in your life causes people to be happier. As the authors of the original study point out: perhaps the result they observed (and we replicated) is explained by lower well-being leading people to limit the diversity of their socialization, rather than the other way around. Or it could be that there is some other factor, say, such as access to fulfilling employment, which increases both their well-being and their opportunity to have diverse social interactions.
And, of course, many other things can influence your well-being; social portfolio diversity is one small factor among many that are related to happiness.
Despite some open questions, it remains true that diversity in one’s social interactions appears to be predictive of happiness for people in the U.S.
It therefore may be worth trying as an experiment in your own life to see if you feel it improves your own well-being. To try diversifying your social interactions, you could:
Seek New Relationships: Regularly engage with a variety of people, such as neighbors, friends, coworkers, and acquaintances. This can be as simple as striking up a conversation with a neighbor or joining a community group.
Reconnect with Old Contacts: Sometimes, we lose touch with people over time. Reaching out to old co-workers or distant relatives can add another layer to our social interactions.
For help building and improving your relationships, see our relationship tools.
We conducted this study as part of our Transparent Replications project by Clearer Thinking. This project aims to greatly increase the probability that new psychology papers in top journals are replicated shortly after publication so that high-quality results are celebrated while non-replicable results are almost immediately uncovered, and researchers have a fast-acting incentive to not publish false positives or misleading results in top journals. If you’d like to support this project or get involved, here’s how: