Work and such

[kryptonitemonkey]

It's so often rough to have a job, or anything really, when the requirements are difficult to the point of impossibility. Working the back end of call center (no selling, just surveys), I am one of two or three others who have the onerous duty of making sure that the actual surveys we get fit the required quotas. Age, ethnicity, gender, area, political affiliation, and occasionally many others, all are numbers that we have to try and manipulate. I say try, because while the client may say we have to get 70 completes in one specific ethnicity, there is zero guarantee we will be able to.

Granted, the numbers that we get all have data associated with them. Most jobs are not randomly dialed at all. Culled from voter registration or sold from pretty much anywhere and everywhere, the vast majority of the numbers we work with are already associated with the required demographic information; meaning we have some granular amount of control. Too many females? we can dial more males, or even only males. Naturally, many numbers are not valid. Half the job of being on the phones is screening out all the answering machines, wrong numbers, disconnected numbers, fax lines, etc. It can take up to 100 numbers to get one complete. But we do have some amount of control.

Sadly, it's still a very organic job, since it's all about people. The people we're calling, the people manning the phones, how each is feeling at the moment of answering, etc. And damned if there isn't always one or two quota groups that WON'T COOPERATE. *headdesk* When a job of hundreds of surveys is closing up and the whole thing is screeching to a halt because you need 20 more of this age, or that ethnicity and, despite running through thousands of numbers and dwo dozen people on it, you have only acquired one in two hours...

Sometimes a job going wonky is my fault, not catching something quickly enough, or mishandling my resources, but just as often it's not. For doing nothing but manipulating numbers all day, my job is oddly organic and fluid in nature. It can be both rewarding and highly frustrating. I left work today, having managed to close up a job that, while not great, at least did okay, with anything wrong with it not my fault. I think I'm tired, because I feel kind of crappy about it. Not terribly so, but more so than is usual for me. I'm normally quite calm, even when shit hits the fan and blows every which way. I tend not to take much of it personally (even when I perhaps should), but today, when it really wasn't a huge deal and no one was upset about it, I am. Maybe I'm just contrary, as I am in so many situations.

Thanks goodness tomorrow is my day off though. I don't even normally work Sundays. They just needed some extra help today. I do wish it payed a little better, especially now that I have to look for a new place to live. My roommate is selling his house and I'm not likely to find such nice accommodations at the price he's been charging me. Making only a buck more than minimum wage (even Cali minimum wage) is perfectly livable when you only have to pay $200-250 a month for rent, but will be much less fun once I move elsewhere. It's been a nice break, living here though, that's for sure. I'll definitely miss having my own washer and dryer.

Apparently I had a lot to say.

Comments

[sparowe]

I can relate, to an extent. Different work entirely, but there's a certain amount of... being unable to get the upper echelons to realise that at the end of the day, you can't make people do anything. Yes, they're required to give notice for X, but what's the penalty if they don't? When the expectation is that we will provide excellent service regardless of circumstances, then you're sending the message that none of your requirements matter. /sigh

[luxorien]

I know this post is about frustration, but I find the technical details of this fascinating. I've always thought the low participation rate on phone surveys must throw them off.

I am curious: if you have to get more of one demographic group to make your sample representative of the population at large, does that mean you end up tossing out responses from people in other groups? How is that decision made?

[kryptonitemonkey]

The way it works is that we have specific quotas every day of how many of each demographic group we need, along with the daily number of overall completes we need. For a larger job, every day we need to get 50-100 completes a day. We have pages and pages of quotas on our system. The system is set up that once you hit the specified amount, no further surveys are allowed to even really start. The data coming from either the number itself, or the first few demographic questions will throw the caller to overquota screen, which just has us tell the person we called that that's all the questions we have, but then we tell the supervisors, who then tell us in the back room, that age 60-69 just closed, or area 11, democrats, females, etc. We then pull out all numbers associated with that specific quota and hope we don't get any more people in that category, with varying success.

For jobs where we ask for a specific name, it's easier to control, but when we're taking whoever answers, you tend to get a lot of overquotas, which typically occur within the first minute or two of a given survey. The hard part comes at the end of the day when you're down to dialing only asian republican women age 40-49 in one of two remaining areas, or any specific combinations therein. A given job has anywhere from one to six or seven types of quota we have to worry about. Every complete falls somewhere in the mishmash of competing quotas.

But a lot of the process is done by the computer. You answer a question a certain way, it automatically kills the survey, and for various reasons. We often get many people who disqualify for whatever reason, be it they're not likely to vote in the next election (so naturally why would they want their opinion?), or they work for an organization that might "bias" the survey. Some of the more heinous push-polls we have to do start off by asking if the person works for the government or the media, and kicks them if they do. They certainly don't want anyone informed or capable of informing others of what is going on...

[luxorien]

That is fascinating. I often wonder how far off these polls are from the results we'd get if every single person answered. Although, I suppose that would confer its own sampling biases...

[kryptonitemonkey]

I've been told that one main reason why our company is doing well when many others aren't is that we're very good at getting the difficult people to answer, meaning we accurately predict elections and such when others may not. Well, as accurate as can be had, given some of the ridiculously biased tripe we have to push some days. The problem is that no one wants to shell out money for unbiased answers when they can try to shape public opinion, both through the survey itself, and through the subsequent "results" shown to the public. "73% of respondents agreed that this candidate is evil (when asked what they thought of him kicking puppies)." Groupthink is always disturbingly effective.