By 
Most of what shows up when you search "anxious attachment in friendships" is the romantic-relationship playbook with the word “partner” swapped for “friend.” That’s not necessarily wrong, the underlying worry, that a friend is pulling away and you are the last to know, is real and common. But it’s a shortcut, and the honest version treats friendship attachment as its own research question, not a leftover of dating advice.
Where “attachment style” actually comes from
Attachment theory started with John Bowlby’s work on infant-caregiver bonds, later tested experimentally by Mary Ainsworth, whose “Strange Situation” procedure sorted infants into secure, anxious, and avoidant categories based on how they responded to a brief separation from a caregiver. That’s the actual origin of the vocabulary you see everywhere now.
The jump to adult relationships came in 1987, when Cindy Hazan and Phillip Shaver published a study in the Journal of Personality and Social Psychology arguing that romantic love could be understood as an attachment process, and mapping Ainsworth’s infant categories onto how adults behave in romantic relationships. That single paper is the ancestor of almost every online attachment-style quiz and nearly all the “anxious attachment” content built since. It was built for romance. Friendship wasn’t the subject.
What the numbers say, and what they don’t
The most commonly cited attachment-style breakdown comes from Kirk Mickelson, Ronald Kessler, and Phillip Shaver’s 1997 study in a nationally representative U.S. sample: 59% of adults classified as securely attached, 25% as avoidant, and 11% as anxious. It’s a real, well-replicated figure. It’s also, like Hazan and Shaver’s original work, built on a romantic-relationship-focused measure, not a friendship-specific one.
Friendship-specific attachment research exists, but it’s smaller and newer. A 2000 study by Chandra Grabill and Kathryn Kerns in Personal Relationships found that people with more secure attachment styles reported more self-disclosure, more responsiveness from their friends, and a stronger sense of being understood and validated within those friendships, compared with less securely attached people. A 2010 study by Ronald Welch and Melissa Houser went further, proposing that friendship attachment deserves its own measurement model rather than borrowing romantic-attachment categories wholesale. Studying 268 college students, they found securely attached people reported significantly more hope, self-disclosure, and relationship satisfaction in their friendships than fearfully attached people did, a pattern distinct enough from the romantic literature to justify treating friendship as its own domain.
Adults classified as securely, avoidantly, and anxiously attached, in the most-cited nationally representative U.S. sample
Mickelson, Kessler & Shaver, J. Personality and Social Psychology, 1997
College students studied in the first model built specifically to measure friendship, not romantic, attachment
Welch & Houser, J. Social and Personal Relationships, 2010
What it actually tends to look like
Set the theory aside for a moment. In practice, anxious attachment in a friendship usually shows up as a handful of specific habits, not one big feeling.
Rereading a text for a tone that probably wasn’t intended. Feeling unsettled by a canceled plan or a reply that came a few hours later than usual, and quietly building a story around it. Asking, directly or indirectly, whether everything is okay between you more often than the friendship actually calls for. And, maybe most costly, staying quiet about a real, specific grievance because raising it feels riskier than just living with it.
If this pattern shows up with more than one person, the fix isn’t the same friend.
Try the NexSpark friendship quiz →Why treating it as a leftover from dating advice gets it wrong
Here is the practical stakes of the Welch and Houser finding: your attachment pattern isn’t necessarily one fixed trait that follows you into every relationship. Someone can be genuinely secure with a romantic partner and anxious with friends, or the reverse, because the two kinds of relationships carry different histories, different stakes, and different unwritten rules. If you took an online attachment quiz built for dating and are now worried about your friendships specifically, you are applying an instrument calibrated for one domain to a different one. That doesn’t make the worry fake. It does mean the diagnosis deserves a second look before you treat it as settled.
What actually helps
Name the specific worry, not the vague feeling. “I am reading a delayed reply as her being upset with me” is something you can check against reality. “I feel anxious about this friendship” isn’t.
Build in a pause before seeking reassurance. A short delay between noticing the worry and acting on it, sending a follow-up text, asking if everything is okay, gives the feeling room to settle before you act on it.
Say the actual need out loud. “I would love to see you more often” lands better and gets answered more honestly than fishing for reassurance through indirect hints.
Spread the weight across more than one friendship. A lot of anxious attachment in friendships gets more intense when one relationship is carrying all of your sense of social security. A Circle of several people, rather than one best friend holding everything, tends to lower the stakes of any single friendship’s normal ups and downs.
If this pattern feels bigger than any one friendship, and closer to a running theme across most of your relationships, that’s a different, deeper question than a single text exchange, and it’s worth working through directly rather than friendship by friendship. NexSpark’s Magnetism track was built for exactly that kind of pattern work.
Methodology & limitations
- Most attachment research covers caregiver-infant bonds and adult romantic relationships, not friendship specifically. The Grabill & Kerns and Welch & Houser studies are genuine, peer-reviewed friendship-attachment research, but the field is far smaller and newer than the romantic-attachment literature, and neither study used a nationally representative sample.
- The 59% / 25% / 11% split (Mickelson, Kessler & Shaver, 1997) is a romantic-relationship-calibrated figure applied nationally, not a friendship-specific breakdown. We found no comparably large, nationally representative study measuring anxious attachment specifically within friendships.
- The Welch & Houser and Grabill & Kerns samples were both college students, which limits how far their specific findings generalize to adults at other life stages.
- We used no NexSpark member data. Every figure on this page comes from published, external research.
Common questions
What is anxious attachment in friendships?
It’s a pattern, carried over from attachment theory, where someone worries their friends will pull away or stop caring, and looks for repeated reassurance that a friendship is still secure. In practice it often shows up as overreading a delayed text, feeling responsible for keeping a friendship afloat single-handedly, or having trouble raising a real complaint for fear it will cost the friendship.
Is anxious attachment the same in friendships as in romantic relationships?
Not necessarily, and most popular content skips this. Attachment theory was extended to adult romantic relationships by Hazan and Shaver in 1987, and the quizzes people usually take to find their attachment style are calibrated on romantic items. A 2010 study by Ronald Welch and Melissa Houser proposed treating friendship attachment as its own construct, since a person's attachment pattern in dating doesn’t automatically predict their pattern in friendship.
How common is anxious attachment?
In the most-cited nationally representative U.S. study, Mickelson, Kessler, and Shaver (1997) classified 59% of adults as securely attached, 25% as avoidant, and 11% as anxious, based on a romantic-relationship-focused measure. There’s no equivalently large, nationally representative study measuring anxious attachment specifically within friendships, so that split should be read as a general reference point, not a friendship-specific figure.
What are the signs of anxious attachment in a friendship?
Common patterns include rereading a friend's texts for hidden meaning, feeling unsettled by a canceled plan or a shorter-than-usual reply, seeking frequent reassurance that the friendship is fine, and avoiding honest conflict out of fear it will end the relationship. A 2000 study by Chandra Grabill and Kathryn Kerns found that people with more secure attachment styles reported more self-disclosure, more responsiveness from friends, and a stronger sense of being understood, suggesting the anxious pattern tends to work against the closeness it’s trying to protect.
What actually helps with anxious attachment in friendships?
Naming the specific worry instead of the vague anxious feeling, building a brief pause before seeking reassurance, communicating a real need directly rather than testing a friend indirectly, and not letting one friendship carry all of your sense of social security all help. Spreading that weight across a few real connections, rather than one, tends to lower the intensity of any single friendship's ups and downs.
Sources
- Cindy Hazan & Phillip Shaver, Romantic Love Conceptualized as an Attachment Process, Journal of Personality and Social Psychology 52(3), 511–524, 1987.
- Kirk R. Mickelson, Ronald C. Kessler & Phillip R. Shaver, Adult Attachment in a Nationally Representative Sample, Journal of Personality and Social Psychology 73(5), 1092–1106, 1997.
- Chandra M. Grabill & Kathryn A. Kerns, Attachment style and intimacy in friendship, Personal Relationships 7(4), 363–378, 2000.
- Ronald D. Welch & Melissa E. Houser, Extending the four-category model of adult attachment: An interpersonal model of friendship attachment, Journal of Social and Personal Relationships 27(3), 351–366, 2010.
Figures are rounded as published. See each source for full methodology. Journalists and writers are welcome to cite this page, just link back so others can find the sources too.