One of the most durable findings in sociology is that people hear about jobs more often through acquaintances than through close friends.
The explanation is that close friends occupy the same social world and therefore know the same things. An acquaintance is connected to a different pool of information, so the value of the tie comes precisely from its weakness.
Large-scale analyses of professional networking data have since supported the mechanism, finding that moderately weak ties are associated with job mobility more than the strongest ties.
This is relevant to sixteen-year-olds for a reason that is rarely made explicit: weak ties take years to accumulate and cannot be assembled at the moment they are needed.
What a network is at that age
Nothing that looks like networking. It is teachers, coaches, family friends, the parents of friends, people met through part-time work, people from clubs and teams, and older students.
These are precisely the weak ties the research is about, and they are being accumulated or not accumulated during exactly the years when nobody thinks of the process as consequential.
By the point at which someone is told to network, at twenty-two, the accumulated stock differs enormously between individuals and is largely a function of the previous decade.
The distribution problem
Weak-tie networks are unevenly distributed, and the unevenness tracks background closely.
A student whose parents work in professional occupations has ambient access to a set of adults in those occupations. She hears how those jobs work, what they are called, and what the entry routes are, without any deliberate effort by anyone.
A student without that access has to acquire the same information deliberately, and generally does not know that it exists to be acquired.
Research on the graduate labour market has repeatedly found that informal contacts and unadvertised opportunities account for a substantial share of entry-level placements, particularly in the most competitive sectors — which means the network is not a supplement to the formal process, it is a parallel one.
Where the gender pattern enters
It is less about the number of ties than about their composition and about how they are used.
Studies of professional networks find differences in structure — with some evidence that women's networks are more likely to be homophilous, that is, composed of people similar to themselves, particularly in male-dominated fields where the available senior contacts are mostly men.
Research on informal contact also finds differences in access to the settings where weak ties form in some industries, though this varies enormously by sector and has changed considerably over time.
The finding with the clearest practical implication is about activation rather than formation: several studies find that women are less likely to ask their contacts for concrete help, including referrals, and more likely to use contacts for information and advice.
A network that is not asked for anything produces less than one that is.
What actually builds ties at that age
Sustained participation in anything with adults in it. Sport, music, volunteering, part-time work, community organisations. The specific activity matters much less than the duration and the presence of adults who are not teachers.
Part-time work is underrated here. It generates references, adult contacts outside the family, and — importantly — practice at professional interaction, which is a skill and not a disposition.
Staying in contact after leaving something. Almost nobody does this at eighteen, and it is the cheapest available action. A coach or supervisor who last heard from you when you left is a much weaker tie than one who received a message a year later.
Asking adults what they do and how they got there. The information yield is high and almost everyone is willing to answer, because the question flatters.
The asking problem, addressed directly
The main obstacle to activating a network is the belief that asking imposes a cost on the person asked.
Research on help-seeking finds that people systematically underestimate others' willingness to help, and overestimate how burdensome the request will seem. This finding has been replicated across several contexts and is one of the more reliable results in the area.
The practical version: the request that feels significant to the asker is usually minor to the recipient, and the base rate of agreement is much higher than expected.
It also matters what is asked for. A vague request is hard to fulfil. A specific one — an introduction to a named person, a look at a document, twenty minutes on a call — is easy to say yes to.
What schools could do and mostly do not
Careers provision is overwhelmingly informational: what jobs exist, what qualifications they need. Very little of it addresses access, which is the binding constraint for the students who need it most.
Structured alumni contact, employer mentoring schemes with sustained rather than one-off contact, and work placements arranged by the school rather than by the family are the interventions that substitute institutionally for what other students get through their parents.
Evaluations of such programmes are mixed and the better ones tend to be intensive rather than broad — the same finding as everywhere else in this field.
But the underlying logic is sound and simple: the students with no weak ties are the ones for whom the school is the only available source of them, and providing them is a redistribution that costs nothing except organisation.