Simple Social Network Analysis Project

Simple SNA Project

My first interest into Social Network Analysis was as a VP of HR, interested in using SNA as a means to help retain key knowledgeworkers during the coming employee shortage caused by aging baby-boomers retiring and the declining birth rates (there are numerous articles on this).

Then there was the next troubling set of statistics that came from three separate reports. Approximately 80% of training budgets get spent on formal training, which accounts for only 20% of corporate learning (to be fair, one report had it at 70/30).

In fact, there is the 75/25 Rule of Learning (studies by Digital Equipment Corp, and another by the US Dept of Human Health and Human Services) which indicates that 25% of what someone needs to know to do his or her job is learned in formal training, the other 75% is from informal sources.

Knowing the social networks of an organization, then, would result in discovering who in the organization is trusted and gone to for knowledge (and therefore devastating to lose).

While there is software and a number of ways to study an organization’s Social Network(s), the simple, and perhaps fast, answer of “how to” in researching your company’s social networks is to survey. There are at least three critical factors that have to be considered before entering into discovering your company’s social networks through the use of a survey. These three factors depend on the type of organization.

1) The section/department/company a wide open, trusting, and completely team oriented group.

2) There some trust and openness in your friendly organization. Or,

3) It is a hierarchical command-and-control company.

The survey can be this simple; simple as asking these questions:

1)                  When you need key information concerning your job, from whom (within the company/department/team/group/section) do you get help?

2)                 When you need help of any sort, from whom (within the company) do you get advice?

In an open organization you should ask for the employee's name on the survey. In a company where there is a fair amount of trust and openness, but there are some barriers in this area, your survey would ask the same questions, except that you have to promise that this is a completely confidential survey. Do not put a place for the employee to supply their name when responding. You only want the data…the names of those who people go to when they need help solving job related problems.

In a command-and-control organization, the dynamics are different, but the social nets still exist. The difficulty is that in actual practice, the questions provided to discover a company’s social nets usually result in the name of a manager, director, or other company executive, as that is the way information flows (I learned a lot from doing this wrong). The command aspect may also dictate that the employees give an obligatory answer of a manager’s name, which may be the trained, reflex response, orientation of where someone would go for information and help.

In order to get past a hierarchical response to the question, the above questions must include a limiter along the lines of, “When you need help of any sort, from whom within the company (other than your manager) do you get advice?” Any and all questions must have this in order to take way the senior/subordinate element out of the response. You may still find that a certain manager is a key holder and sharer of knowledge, but you will get different results if you run the survey first without the question limiter, and then adding the limiter afterward.

Once the survey is done, tabulate the results, and it will be evident who the real knowledgeworkers are and where employees go for key information to do their job and assist the organization.

In “high trust” organizations, you can get an exact chart of who goes to whom for help and advice. In an organization with less trusting lines of communication, you can at least get the most helpful and trusted employees as a vote of confidence.

Comments

[Michael Jones]

Interesting observations and analysis of the sociograph. I've found in my research that sociographs are quite handy to relay to the organization what the informal network looks like in practice. Members of the Cornell FSAE team were quick to pick up on these structures and relations.

I found the note about formal budget allotments for KM being the opposite of actual learning practice quite interesting - I always strongly suspected this to be the case but didn't have a source to back it up.

I would argue it's partially a return on investment issue - informal learning is cheaper (but more complicated) to facilitate, whereby proper formal learning requires a lot more investment to pay off.

This isn't to say it's not useful or necessarily inefficient - for some learning problems, it's probably the most efficient solution and worth the money.

But many problems - specifically those for which tacit knowledge transfer is essential to derive a proper resolution - are probably better and more efficiently dealt with by stitching together a series of informal learning mechanisms.

Again, something I was able to learn from car team. We never had the luxury of having enough money or human capital to sink into formal systems anyway, so informal learning mechanisms became that much more important, and it's what I'd spend most of my time on.

Unfortunately, formal systems and models make people feel more comfortable. Much of my dissertation research doesn't cover a lot of the qualitative informal learning solutions I helped advance, because my committee felt the need for numbers.

So, I met their needs but continued to do the right thing on the side anyway. :-)

Thanks for dropping by my slice of the blogosphere and leaving some insightful comments. I'll be sure to drop by here more often now that I have you in my RSS reader.