Two reports on the US loneliness About one released in 2025 41 percent of adults were single. Another said about 61 percent. Same country, same year – 20 percent difference.
This was not further research. These were experts who used built-in measuring instruments. And yet they painted radically different pictures of the same population. If the numbers vary so much depending on which questionnaire you take, what are we looking at?
My colleagues and I have studied this issue in depth LONELY-EU projectAn EU-funded effort to assess how loneliness is measured in all 27 EU member states. We have recently published two political references – one Europe’s measurement problem and one quality assessment system we’ve developed a solution to it – and I want to share what we’ve found because it has implications beyond the EU.
Ruler problem
Loneliness is associated with a The risk of premature death increased by 26 percent. In Spain alone, it is expensive more than 14 billion euros per year. The Great Britain and Japan appointed ministers of solitude. It was said by the former surgeon general of the United States public health crisis (but see this post as a reminder that it is not an epidemic). The World Health Organization was launched Commission on social relations. There is real political power here. But as we lay in ourselves Policy Brief Statement of Measurement Issuesthe instruments we use to measure loneliness are not designed for what we now ask of them.
When our team evaluates basic loneliness questionnaires Surveying more than 25,000 people in all 27 EU countries, we encountered a paradox. One measures it possible meaningfully comparable across countries (the three-item UCLA Loneliness Scale) covers mostly painful items. feeling from loneliness. It is weaker in determining whether someone is truly lacking in social relationships or facing systemic barriers to communication. Thus, a program that successfully connects isolated people may appear to have failed—not because it didn’t work, but because the tool failed to see what had changed.
The measure of this does get the full picture – emotional distress and social support (De Jong Gierveld scale) – can be reliably compared in only 5 of the 27 EU countries. Use it to compare loneliness in France to Poland, and you might be comparing an apple to something that isn’t even a fruit. And then there’s the single-question approach that many surveys still rely on (“How often do you feel lonely?”). Our data show that they do not pass the validity test in eight countries and may not reflect the multidimensional nature of loneliness at all.
This is not just a problem of methods
An elderly widow living in rural Greece experiences loneliness differently than a young professional in Amsterdam or a refugee in Berlin. Current instruments, developed primarily in the United States and the Netherlands, may not capture these changes. If the measure fails to “see” certain forms of loneliness, the entire population is left out of the data. And populations that don’t show up in the data won’t have resources.
In the same EU country, prevalence estimates from different surveys in the same year can vary by 8 percent or more. When the numbers vary so much from survey to survey, try to gauge whether the national loneliness strategy is working.
Traffic light system
Ours Acronym for measurement quality policywe are introducing a traffic light rating system to help solve this problem. It evaluates each dimension of loneliness on three dimensions: Can you compare scores across countries? Does it really measure loneliness (it doesn’t). depression or general dissatisfaction)? And does it fully capture the experience of loneliness?
Green means strong evidence. Yellow means caution. Red means you don’t use it to make decisions.
No measure available will have green in all three. The UCLA scale scores green for comparability and validity, but yellow for content coverage. The De Jong Gierveld scale scores green for validity and content, but red for comparability. Single element scale: red, yellow, red.
The picture is exciting, but at least it’s honest. If you’re a policymaker trying to decide how to track loneliness across the continent, you need to know what your tools can and can’t do.
What’s next
We are developing the EU SIL Index, a new measure designed to achieve what no current tool can: reliable cross-country comparisons and comprehensive content coverage across the broader domains of social communication, set in multiple EU countries and translated into multiple EU languages.
Read about loneliness
But a better poll won’t get us there. Europe also needs the infrastructure to put it to good use: coordinated data collection, quality assurance protocols and integration with statistical offices. You don’t just need a good test—you need a system to administer it, collect results, and act on them.
We know the importance of social communication. Our ability to track it, compare it across populations, and determine whether our interventions are actually working depends on getting the measurement right. It’s not the most engaging part of the conversation, but I’d argue it’s the most consequential.
