People’s personal values change in predictable ways as they age, and examining these values at a more granular level can reveal much more about a person’s stage in life than broader measures alone, a new study published in 2016 finds. Personality and Social Psychology Bulletin.
Psychologists have known for decades that a person’s personality changes over the course of a person’s life. Young people tend to focus on excitement, joy, and personal accomplishment, while older people typically value safety, tradition, and social harmony more. Although these trends are generally well established, most studies only examine values at a broad level and group them into large categories that can obscure important underlying complexities.
The researchers behind this new study wanted to see whether a closer look at individual values (such as curiosity, risk-taking, and helping others) could reveal patterns missed by broader groups. They also wanted to test whether these more detailed values could more accurately capture how people change over the lifespan, compared to the broader categories typically used in psychology.
The team, led by Andres Gvirz of King’s College London, analyzed data from 80,814 people aged 18 to 75 (57.4% women, 35.8% men) collected through an online survey hosted by Time magazine between 2017 and 2023. Each person completed a 20-question survey called the 20-Item Values Inventory, which measured how much they identified with specific principles, such as being creative and being inquisitive. Having fun, respecting authority, or caring for others.
The researchers then combined traditional statistics with machine learning models to see how accurately these values could predict a person’s age. They compared results across three levels: four broad higher-order value categories (such as “openness to change”), 10 mid-level fundamental values (such as “self-direction”), and 20 highly specific value nuances (such as “curiosity” and “creativity”).
The results confirm that people’s values change significantly as they age, but the full story is only revealed when observed at the most detailed level. At the broadest level, the data showed a familiar pattern. Older people value conservation and stability more, while younger people lean more toward openness and self-enhancement.
But when researchers zoomed in further, a more complex and sometimes contradictory pattern emerged. For example, within the basic value of “compatibility”, respect Cherish it and it will increase with age. act appropriately It actually decreased. Similarly, for “charity,” older participants placed more importance on aligning with the charity. needs of othersbut less actively done help people. These opposing trends canceled each other out on a broader level, with nuances becoming invisible only upon closer examination.
Detailed value items also proved to be significantly better at predicting age. The broad value categories explained approximately 4% of the variation in participant age, whereas the most specific value items explained approximately 12%. This is approximately 3x more predictive power. The researchers found that if the participants were at least 20 years apart in age, the computer model could accurately guess which of two randomly selected participants was older using only 20 specific value questions with an 80% success rate.
As the researchers note, aggregating specific values into broader categories “misses important information, produces inconsistent results when nuanced, and significantly reduces predictive power.”
This study is not without limitations. Because this is a cross-sectional study, meaning we compare different people at different ages rather than following the same individuals over time, we cannot rule out the possibility that some differences reflect generation rather than developmental change. For example, someone who is currently in their 60s may have grown up in a different social and economic era, which may have shaped their values independently of their biological age.
Additionally, this dataset is heavily biased toward younger U.S. populations, and findings may not be fully generalizable to older or non-Western populations.
The study, “Human Values Across the Lifespan: Age Differences at Three Hierarchical Levels and What We Can Learn from them,” was authored by Andres Gwirtz, Matteo Montecchi, Amy Selby, and Friedrich M. Goetz.
