Survey--Success!

We reached our goal of 300 respondents. Hooray, and we're all glad to be done with it. Now we just have to finish inputting, and analyze the data. The inputting will go fast, but with the end of the school year coming up, and a lot of writing deadlines I have, I doubt I'll get much beyond looking at frequency distributions before next fall.

At another level of the research, we toured a 1400 cow dairy CAFO yesterday. A fascinating operation--from the mortality compost pile (dead calve mixed with hay and manure) to the automatic milkers that pop off when the milk flow drops below 1/2 gallon per minute.

The farmer said the cows probably average 30 gallons each per milking (they measure by weight and by groups of milkers, rather than by individual cow), and are milked 3 times a day. So I estimate, roughly, 90 gallons per day over a 5 year milking period (after that they magically turn into hamburgers on my grill) equals 164,000 gallons per cow.

And the damn things eat better than you do, too. They eat a scientifically developed feed that is very nutritious, and which I can only assume costs a small fortune.

And, interestingly, in contrast to what I have been led to believe, the cattle were not chained up and stuck in one position. They were in large pens that allowed them to walk around, and had raised beds of sand that allowed them to lay down out of the muck, as well as access to water in raised troughs (that keep the muck out of the water). And the muck is squeegeed out during their milking each day.

Here's the part the economist inside me liked most: the squeegees are made out of old tires from earthmoving equipment. Some company in Michigan buys the old tires, cuts them into sections, and puts connectors that allow them to be attached to bobcats. A good case of the market using resources well, rather than wasting them.

Survey--Day 3: Still Getting Better

Tonight we had 70 responses, bringing our total up to 170. I recalculated the number we need, and decided that we could get a 95% confidence rate by dropping from plus/minus 5% to plus/minus 6%, with 266 responses. I want a few extra, but we're at least looking at a succesful survey now, with maybe two day's more work. I estimate we have about 45 man hours on the phones now, which works out to about 3 3/4 responses per man hour--a lot of work for little payoff.

Day 2: Better

Well, we had better luck today. We got 60 responses, so our two-day total is 102. Still less than halfway there, but if tomorrow night goes well, and Saturday goes well, we might be close enough that the students are ready to make a final push for 271.

Obviously I haven't looked at the data systematically yet, but on first impressions, there seems to be a good split between strong critics of CAFOs, strong defenders of them, and a lot of people in the middle. Interestingly, very few people seem to recognize the term "CAFO," until we explain it, which might be an indicator that feelings generally don't run too deep on the subject.

My personal favorite tonight was a person who is very strongly opposed to CAFOs, worried about harm to the environment, thought they should be regulated more strictly, but self-identified as "very conservative." As that's the last question, I was primed for a very different response. And that's what makes social science difficult--people don't always behave in individually predictable ways.

Bad Start to Poll

We need at least 271 responses to our survey. First night....42. I have 10 very discouraged students, and can't say I'm not discouraged myself. We'll do tonight and tomorrow night, and I think most of the students are agreed to work on Saturday, as well. If we can get close to our goal, maybe they'll be encouraged to come back for another day or two to finish up.

I have to say, I've earned new respect for honest people. It's much better to have someone simply say, "No, I prefer not to," when asked if they will participate, than to have someone just hang up on you. My personal favorite was the person who said yes, but needed to put me on hold for about 30 seconds. I was already wondering if it was a trick when the line went dead after about 10 seconds.

Never having done survey research, I was intrigued to notice that (a) it was very hard to stick to the script, and (b) a substantial number of respondents don't like to stick to the closed set of responses. Question: "Some people say CAFOs harm the environment. Do you strongly agree, agree, disagree, or strongly disagree?" Answer: "Well, the real problem with CAFOS is...." I suppose that's human nature, but it makes the survey researcher's job miserable.

I think I'll stick to what I do best--digging up numbers, facts, and figures, and pretending average citizens don't exist and don't matter when it comes to political research.

CAFO Survey This Week

Starting tomorrow my Research Methods class is starting its survey of county residents about CAFOs (Concentrated Animal Feeding Operations). In our county they're mostly dairy farms run by Dutch immigrants, with one hog farm, and each operation having between 700 and 1500 animals.

They're very controversial, as they allegedly have difficulty managing the waste in a way that keeps it out of our rivers. The river that flows through our town (1 block from my house) has elevated levels of E. coli. This mostly comes from spraying semi-liquid waste on the fields, which has the dual benefit of allowing the CAFOs to get rid of it and providing crucial nitrogen for growing crops, but the drawback of sometimes draining into surface water. They're also controversial because Americans are still in love with the myth of the pastoral family farm, and hate "factory farming," or "corporate agriculture." The anti-CAFO forces are quite vocal around here, but it's not clear that it's more than a handful of activists, especially as this is a very conservative county.

But starting tomorrow, we'll find out what the average county residents think. Besides allowing me to have research assistants for something I'm interested in, it gives the students a chance to experience how frustrating real data collection can be, but also gives them a chance to write up the results and take them to a conference, or try to get them published--either of which would look very good on a grad school application.

But I've never done survey research before, and I really don't like talking on the telephone that much, especially to strangers. Other than ordering pizza, I usually let my wife make the business calls. So it could be a very stressful experience, although I hope not.

I'll post bits of results when we get the data entered.