Saturday, September 18, 2010

The Strength of “Work” Ties

Most people interested in Social Networking, whether professionally, academically or as a user, are familiar with the concept, “the strength of weak ties”. The paper of the same name by Mark S. Granovetter of Johns Hopkins University issued originally in the early 70’s and updated in 1983, analyzing social networking has been very influential in the field and I believe, is responsible for some miss-conceptions regarding how jobs are found.

Although not specifically stated in Granovetter’s paper the generally accepted conclusion on job mobility reached by many, is the following; most people find jobs through people with whom they don’t have a strong relationship.

Having been a recruiter for almost twenty years, experience tells me that this is contrary to what actually occurs when an individual changes jobs. In this opinion article I endeavor to uncover a new perspective. I will also identify problems with the methodology that led to conclusions that do not reflect practicality, as well as use real world examples to support the counter-argument that it is, The Strength of “Work” Ties, that facilitates mobility in the job market.

Labor Market Studies and Empirical Data

Once the concept of the “Strength of Weak Ties” had developed, empirical data was gathered to further develop the cohesive power of weak ties. One of the underlying suppositions for the labor market study was the idea that those with strong ties to the job seeker would be more motivated to help with the job search.

In 1970 Granovetter studied how people in professional, technical and managerial positions found their new jobs. Particular emphasis was placed on the nature of the tie between the job changer and the source of the employment opportunity. By asking a random sample of one hundred individuals who found jobs through contacts, how often they saw the contact around the time of the job change, and using contact frequency to determine tie strength, he was able to offer some conclusions.

The results:

Sample Size 100 individuals

% of Sample Tie Strength Contact Frequency Definition

16.7 % strong ties = often=at least twice a week
55.6% occasionally => once a year but <>
27.8% weak ties rarely= once a year or less

From these results he concluded, ”the skew is clearly to the weak end of the continuum, suggesting the primacy of structure over motivation.” Or in other words, most of the individuals found jobs from weak ties and strong ties are not as important as weak ties for finding a job.

Heuristics are the Problem

First of all, a study with such impact on the thinking regarding job mobility seems like it should be based upon more than a sample of only one hundred individuals.

This opinion essay recasts the study with a focus on professional networking related to job change. Granovetter’s use of “professional networking” examples to reflect social networking phenomena creates one of the basic problems with the SWT study. Proximity has been identified as a factor when establishing strong ties in a social network. It makes sense that people that see each other more frequently have a higher probability for forging a strong tie. Movement within social networks tends to be gradual unless one uproots. In a social network, once proximity is not a factor the ties diminish in strength. I argue that Professional networks are different because movement is abrupt from one company to another yet the strong ties remain in spite of lack of contact and proximity.

Another factor that makes professional networking much different from social networking is that much more is at stake. When you are a reference for someone, your reputation will be affected by that reference and therefore has more consequence than just passing on information. I think we can all agree that how someone finds a new job is not exactly the same as how the rumor of Sinbad’s death diffuses through a social network.

Definition of Strength

Another problem is that the definition of a strong tie is applied arbitrarily and the resulting determination is the exact opposite of what a strong tie is in a professional networking context.
First, let’s start with the problematic method used in the study to define the “strength” of a tie. Granovetter’s definition initially uses the concept of time, frequency of contact, emotional intensity, mutual confiding and reciprocal services, as criteria to determine the strength of a tie. Then for no apparent reason in the labor market study, which is responsible for most of the data used to support his arguments, uses only the contact frequency to determine the strength of ties. As we will see, in a professional networking scenario, frequency of contact does not determine a strong tie. Therefore, strong ties may be mistakenly defined as weak, which produces incorrect conclusions.. In general my intuitive notions do not agree with the criteria related to time and contact frequency with respect to tie strength. I have close friends that I see or contact very infrequently, yet they remain strong ties.

I propose it may be much easier to define and determine the strength of ties in a professional networking context. In this analysis, I am using referral as the defining criteria. For example, would individual A refer B? If so, a strong tie exists. Furthermore, using the concept of transitivity, if A has strong ties to C and refers B to C, we will infer that B has a strong tie to C. A weak tie occurs when A knows B but does not refer them. This is not necessarily a negative; it may also be due to lack of sufficient knowledge about B’s performance etc.. Staying true to the comparison to Granovetter’s paper, I will stick with the two qualities of ties, weak and strong. However, our experience reveals the following data points of interest. If we continue outward to the next degree of connectivity we find that the strength of the tie attenuates somewhat and that beyond three degrees does not register. The significant data point being that a referral of D from A to B to C is less than a strong tie, but stronger than a weak tie.

Common sense and experience tells us that time spent with an individual and frequency of contact are not much of an issue when determining strength of a tie based upon referrals. Certainly a reference of A for B from ten years ago is not as strong as one currently, but taken in context between strong ties A-C, it is still considered a strong tie for C -B.

Lack of contact with a former colleague for a few years, does not diminish the strength of your referral. For example, someone that worked with Ray Ozzie four years ago, yet has had no contact since, will likely give just as strong a referral as someone with whom Ray works currently. As a corollary, the best people you work with are not necessarily the ones you frequently hang out with, and some might suggest that the amount of time one spends around the “water cooler” is inversely proportional to the quality of the individual.

The remaining criteria, emotional intensity, mutual confiding and reciprocal services would still apply when using referencing to determine the strength of tie in a professional context.

The “Strength in Numbers” of Weak Ties

An obvious explanation for the preponderance of weak ties job changes would be the greater number of weak ties. If your five close friends know of three jobs between them, but their friends and friends of friends know of thirty, then the probability is good that the job will be found through weak ties. Certainly this is more of a factor when a job must be found in a short amount of time. A footnote in Granovetter’s paper notes that David Light brought this mathematical fact to his attention. However, Granovetter maintains that the objection remains inconclusive, and that “if the premise were correct, however, one might still expect that greater motivation of close friends would overcome their being outnumbered.” As we proceed, I believe that the analysis will substantiate David Light’s suggestion regarding “the strength in numbers of weak ties” and also that strong ties do overcome weak ties when changing jobs.

The Strength of Weak Ties Concept does not reflect reality

Conservative Industry estimates indicate that in the US there are at least 50 million hiring situations annually and 40% of these hires are made through referrals.

When you refer someone, you have skin in the game. Your reputation is on the line and therefore you do not make it lightly. Chances are you are not going to refer someone you do not know well.

This one industry statistic alone is probably enough to convince you that the conclusions drawn from asking the wrong question to one hundred job changers may not be more accurate than data derived from the more than 50 million job changes annually.

An Alternative View

If we reframe Granovetter’s labor market study by asking the job changers the more relevant question, “Was the job source referring them?” we might see different results. Examining the amount of contact we currently have with our references from past jobs, probably most would fall into the once a year or less category. Yet when looking for a new position, the first step is to contact your past network of references and referrals.

Ask any hiring manager what their process is to find and hire people and you will get similar answers. The first step is to vet their network and the network of their employees. If they utilize other recruiting sources and a viable candidate has been identified, backdoor referencing is done. In essence, using these unofficial references, the manager is attempting to bring the candidate into their network of strong ties. Managers try to avoid making a hiring decision based solely on interviews and official references. It is much easier to hire a candidate verified through your network.

Think back to your ten friends that have most recently changed jobs. How did they find their jobs? Chances are, they found the job using their network of work contacts that are references or would refer them. Many times I have seen a past manager that was called to be a reference, make an offer to the candidate seeking the reference. In some cases, B’s referral from contact A to A’s reference C, led B to a job with C. As I have noted, due to transitivity, that would still be considered a strong tie move.

Granovetter said it best, “In many cases, the contact was someone only marginally included in the current network of contacts, such as an old college friend or a former work mate of employer, with whom sporadic contact had been maintained (Granovetter 1970, pp. 76-80). For work-related ties, respondents almost invariably said that they never saw the person in a non-work context. It is remarkable that people receive crucial information from individuals whose very existence they have forgotten.”

How true. People will make referrals even when they have not been in frequent contact. In my experience it is not remarkable, in fact I have had managers hire the same person from me several times due to losing contact with the individual. There is no stronger tie in a professional network than when a manager hires someone again after moving to another company.


In my opinion, The Strength of Work Ties is the major factor in job mobility. References and Referrals are the most powerful conduit for finding and landing a new job.

Opportunities found using this method have a higher probability of fit due to the similarity that exists between individuals with strong ties. This similarity will include the quality of the individuals work output. A common phrase in the recruiting industry is that A’s refer A’s. The likelihood of being hired through strong work ties is increased due to the best fit, the motivation of the internal reference and the lessened risk provided to the manager.
For the same reason, positions found through strong work ties have a higher probability of job satisfaction and longevity.

Respectfully, The Strength of Weak Ties, does play a role in job mobility, but the value lies in the larger number of opportunities available to the individual seeking a new position. This is especially true when there is a lack of references or the individual is pressed for time to make a move.

1 comment:

  1. It’s unclear what the winner of the Miss Universe pageant receives, however it’s rumored to incorporate a six-figure wage, a... On a spring evening in 1968, he drove his white Rolls Royce to the playing den and, over 3 days, proceeded to win $48,000 ($360,000 today). He’d spent hundreds of hours devising an ingenious technique 카지노 사이트 of winning — and it would soon internet him the modern equal of more than $8,000,000. The 0, with its green background, has black numbers on both aspect, and the 00, additionally on green, has purple numbers on both aspect. Using zero as beginning point|a place to begin} on both aspect, the numbers alternate black-red until they reach zero.