Psychology: I tell my students to remember that every line of their data is a person. An actual person with a life outside the tiny slice they dedicated to telling us about. I have always tried to remind myself of this, but it was made real when I was doing my PhD in the year... mumble mumble. That research was based on a longitudinal study surveying the same people four times over the six months before and six months after a general election.
Back then, I would send out hardcopy surveys in the mail and open them as they came back. I remember opening a time-two return to find the survey was blank but for a spidery note written on the front. The writer was sorry, but the recipient of the survey (who had been kind enough to respond at time one) had died. This wasn’t the only occasion I received such a response during that research.
People die, and the least important thing about that is you don’t get any more data.
But longitudinal research projects like this can tell us a lot about people’s lives in the lead-up to their death. I imagine that other prospective participants also died without my knowing, but I could have found out if I wanted to, and maybe I could predict who was most likely to die during that research.
This is what other researchers have done, usually people working on much longer longitudinal studies than mine. For example, in 2008, Antonio Terracciano, now in the department of geriatrics at Florida State University, used information from the Baltimore Longitudinal Study, which started in 1958, to see if it could predict longevity.
During the study, 2359 of the participants completed a measure of personality – the Guilford-Zimmerman Temperament Survey. I’d never heard of this so looked it up. It’s described as a “relatively old” test, designed in 1948, and it doesn’t capture all of the things that we would assess in the 2020s. It also includes 300 questions; I’m definitely not using it.
Of the Baltimore study participants who completed the test, 40% had died by the time Terracciano and colleagues looked at the data. But to the punchline: scores on some of the facets of the temperament test predicted “mortality”. Mortality doesn’t just mean death; it’s also a technical term for a statistical tool called survival analysis. This is what you use when you’re trying to predict an outcome that has either happened or will happen at some point, and we call that outcome “mortality” even when it doesn’t literally involve death. For example, return of a cancer after remission, or reconviction and sentence to prison.
Personality prospectively predicts how long participants lived. For every one point increase in scores for aspects of extroversion, emotional stability and conscientiousness, participants lived an average of 2-3 years longer than people averagely extrovert, stable and conscientious.
Another way to do this research is to find people who have lived an awfully long time and see what is different about them. One such study focused on 548 Utah centenarians. Utah is the home of The Church of Jesus Christ of Latter-day Saints, the Mormons. Mormons tend to live significantly longer than the average American because they avoid alcohol and drugs (and coffee!)
If you make it to 100, this research says, you’ll live even longer if you’re social and close to the people around you, are satisfied with your life, and … get to sleep easily.
The take home from these studies is two-fold. Longevity is about medical and physical health, as well as mental characteristics, and the commonality in these two studies and others is that being with people keeps us alive.
Time to say “Hi” to friends and family.
