A couple of weeks ago I was talking personality with a group of Emergency Managers. They’d recently completed a five factor questionnaire for me as part of a week-long residential programme they were attending. We were discussing how this information could feed into their work in teams, with new teams they might suddenly be thrust into, what they felt about the results, what it all means, and so on.
What a great session and what a wonderful group of people who are incredibly committed to their work. If you were there, I promised to write something, so please share!
Inevitably, other personality scales came up in conversation, with the question of how much we could trust them. Some common four letter or colour ones may have been mentioned. Answer, not much. Which Hogwarts house are you is at about the same level.
One thing that didn’t come up, which surprised me, was birth order.
Let me explain.
What is the birth order idea?
The theory says siblings compete for their parents’ attention, each takes a different role to get it, and the role subsequently solidifies into adult personality. You may have heard of the bossy firstborn, the peacemaking middle child and the reckless youngest.
For what it’s worth, as a middle child, I’ve always found the idea that middle kids are the most awesome pretty apt.
However, confirmation bias aside…
Does birth order predict personality?
Barely.
The largest study finds differences too small to use on one person.
Rohrer, Egloff and Schmukle (2015) combined national panels from the United States, Great Britain and Germany, 20,186 people in all. They found no birth-order effect on extraversion, emotional stability, agreeableness, conscientiousness or imagination. Among the personality measures the one exception was self-reported intellect, which fell by about a tenth of a standard deviation with later birth position. Measured intelligence showed the expected birth-order effect followed suit, falling a teeny bit with successive kids. That would suggest later kids are, on average, a touch less intelligent than earlier ones.
For what it’s worth, as a middle child, I hate this, even if the difference is tiny.
A larger study has since found more. Ashton and Lee (2025) analysed a longer personality questionnaire completed online by more than 700,000 adults, and a second sample of about 75,000 also reported how many children they grew up with. They suspect the earlier scales, two to ten items each, were too brief and too narrow to pick up the differences. In their data, people from bigger families scored higher on honesty-humility and agreeableness. Their agreeableness scale measures patience against anger and is a different scale from the one Rohrer used. The largest family-size gap they report sits between only children and people from families of six or more, and the authors spell out what it means. Take one person at random from each group and the one from the large family is the more agreeable of the two 60% of the time, against 50% for a coin toss.
Birth order itself did less. Once family size was held constant, the gaps between oldest, middle and youngest were about a tenth of a standard deviation at most. If we re-run the coin toss on a gap that size, we get 53% at most, meaning that the youngest might score higher on honesty-humility 53% of the time. So three more per hundred than a coin toss, for honesty-humility. Even then, you only see that gap in really large samples.
So, no, birth order isn’t a thing for personality.
Now what?
Where this gets tricky, is when we look to use tools like this, or some other tools that are much less reliable, and make decisions from them.
Selection is one use of a personality measure, but it’s risky move. Although I hope it’s obvious, methods built around the job itself predict performance best. Start by asking how your team was put together. If it was balanced using a personality test that sorts people into four colours or four letters, or a hunch that someone is “a detail person”, or “right-brained” or “Hufflepuff”, ask what that label predicts about the work.
Then start again.
Sackett and colleagues (2022) revised the validity estimates in Schmidt and Hunter’s 1998 review of selection methods, and their paper sets out what the revision means in practice. “Our findings indicate that the predictors at the top of our list in terms of criterion-related validity are those specific to individual jobs, such as structured interviews, job knowledge tests, work sample tests, and empirically-keyed biodata.” They continue, “These tend to fare better than more general measures of psychological constructs, such as measures in the ability and personality domain” (Sackett et al., 2023).
But what about how the team functions together?
The more common use is the team day, where everyone gets a profile and compares notes, much as we were doing with our questionnaire and session. Colours and letters add nothing there, for the same reason birth order doesn’t. Measured traits are a different matter, and the research on them gives both the person, and a team leader, something to work with.
This was the basis of our discussion.
Two key traits
Two traits stand out.
Bell (2007) pooled field studies of working teams and reported that “team minimum agreeableness and team mean conscientiousness” (as in statistical mean, not nasty mean) were among the strong predictors of team performance. The word ‘minimum ‘is helpful. The least agreeable person in a team told Bell more about its performance than the team average did.
An earlier meta-analysis found the same two traits, and found that teams performed worse when members differed widely in conscientiousness (Peeters et al., 2006). A 2021 study of 104 start-up teams reported that high performance needed every member to reach a minimum level of agreeableness, emotional stability and conscientiousness (Kollmann et al., 2021).
That’s incredibly helpful. Side note, and my view: should you keep toxic high performers? Answer: No.
Here’s what to do with it
What it does do is give a trait measure a job in a team. Choose a scale that reports each trait as a score along a range, with published reliability figures. (We used the BFI-2, with permission from the author to use it in our group.)
Let each person see their own scores, then spend the session on two questions. Where do we differ most on how thoroughly and how punctually work gets done, and what standard will we all hold? How will we raise disagreement so that the bluntest person in the room doesn’t set the tone? Write the answers down as team rules and review them after your next exercise.
The scores stay with the people they describe. They show where a team needs an agreed standard. They don’t show who should hold which role, and Bell found personality effects were negligible in laboratory teams, so expect them to count most in teams that work together over months.
When you’re building a team, build skills. Then figure out how best the different traits work together. Remember that minimum agreeableness and the average of the team’s conscientiousness are strong factors for success.
Photo by Hannah Busing on Unsplash