Please help with a science survey

A colleague at the Center for Biodiversity and Conservation Studies at Loma Linda has asked me to pass this on to you:

IMPORTANT SURVEY ON ATTITUDES TOWARD THE ENVIRONMENT AND CONSERVATION - Your help is needed! The Loma Linda University Center for Biodiversity and Conservation Studies is conducting a groundbreaking survey on attitudes toward the environment and conservation, particularly of those who love plants and animals. The results from this study, to be published in a professional journal, will contribute to our understanding of the role of plants and animals in society. Participants are urgently needed to complete the survey, which should take about 5-10 minutes of your time. Please click on this link.

More like this

Early Human Populations Evolved Separately For 100,000 Years: A team of Genographic researchers and their collaborators have published the most extensive survey to date of African mitochondrial DNA (mtDNA). Over 600 complete mtDNA genomes from indigenous populations across the continent were…
As presidential politics ramp up and the environment becomes part of the rhetoric, it reminded me of a essay I wrote a couple of years ago as part of book project entitled Thirty-five Years Since Earth Day: Visions of a New Generation. The editor ended up dropping the ball on the book, but my essay…
Shift Work Linked To Organ Disease, Study Suggests: Disruption of an individual's natural sleep-wake cycle has been determined to be a contributing factor in the development of organ disease. The findings of U of T researchers were recently published in the Journal of American Physiology. Mass…
There are 62 articles published this week in PLoS ONE. There are also two Journal Clubs going on right now - here and here. Here are some of my picks for the week - go read, rate, comment and send trackbacks: A Comparison of Wood Density between Classical Cremonese and Modern Violins: Classical…

Your colleague needs practice making surveys. Far too many of the rankings on that one were phrased in the negative:

"X is not the case": agree, neutral disagree.

Feh. 29% of the people answering that will get the sense wrong and give an answer false to what they intend. Which makes the results untrustworthy.

I didn't have any trouble understanding the questions.

There were some questions where there were not enough choices, forcing me to skip the question entirely. Not all of the yes/no questions necessarily had a yes/no answer.

I agree with both Russell and Gwen. Unfortunately the questions disappear when completed and one cannot remember the exact wording, but double negatives are always tricky. Wyatt said he had no trouble. I wonder whether he really did understand the questions?
I spent some ten years of my life writing psychometric software and implementing tests and found that the precise wording of a question had an inordinate effect on test results, particularly when conducting them with less-well educated respondees (into which class one might venture to place the majority of the U.S. public?).

I did understand the questions. The "agree/disagree" wording makes it quite simple. The reason "agree/disagree" is used is to avoid the double negative wording.

"There are no birds flying out of my ears"

I think about that. I think about what it means. I strongly feel that there are no birds flying out of my ears. I feel that this statement, "There are no birds flying out of my ears", very much reflects what I am thinking. I thus strongly agree with it.

For this:

"Wyatt said he had no trouble. I wonder whether he really did understand the questions?"

I think, no, I really did understand the questions. There was no time when I did not feel that I understood the statements being made and I had great confidence in picking out an answer reflecting my thoughts. So, I would say, I "strongly disagree" with this statement wondering if I understood the questions.

I agree with Gwen - some of the questions were of the type:

Who created the universe?
(a) Abraham's god
(b) Brahma

As for how questions are asked (as pointed out by Clam#5), that's pretty obvious when it comes to political polls. In the British comedy "Yes, Prime Minister" one character demonstrates the art of crafting poll questions.

By MadScientist (not verified) on 18 Oct 2010 #permalink

MadScientist, I just took the survey, and I did not see any absurd questions like that. Certainly not that one! Which questions were they? Could be the survey changes for different takers?

By P. Locans (not verified) on 18 Oct 2010 #permalink

Well, I don't agree with those who called all the questions into question. On the other hand, I felt that my personal participation had more to do with establishing error bars than influencing actual numerical trends.

Perhaps I would be better at answering surveys if I knew even less about statistics?

You do realise that the sample selection method being used makes this survey valueless don't you?


By CaptainBlack (not verified) on 18 Oct 2010 #permalink

This requires pharyngulation. What an appalling piece of irritating nonsense. i want 3 minutes back

Have no fear, Christine O'Donnell is here to lecture on both science and constitutional law:O'Donnell questions separation of church, state

WILMINGTON, Del. (AP) â Republican Senate nominee Christine O'Donnell of Delaware on Tuesday questioned whether the U.S. Constitution calls for a separation of church and state, appearing to disagree or not know that the First Amendment bars the government from establishing religion.
During the exchange, she said Coons' views on creationism showed that he believes in big-government mandates.
"Talk about imposing your beliefs on the local schools," she said. "You've just proved how little you know not just about constitutional law but about the theory of evolution."

By Virgil Samms (not verified) on 19 Oct 2010 #permalink

Earlier in the debate, O'Donnell accused Coons of constitutional ignorance, saying that "perhaps they didn't teach you Constitutional law at Yale Divinity School."

By Virgil Samms (not verified) on 19 Oct 2010 #permalink

This is "how to pharangulate a survey"? Somehow the readership of Greg Laden isn't quite the typical cross section of society.

I'm going with the assumption that the scientists doing this survey know that they are surveying the Internet and have a sense of that that means. Most projects like this have multiple methodologies. There is a good chance that this has been thought through.

I'm promised a copy of the resulting publication. We can check out how it all went!

If we assume that those behind the survey are not total incompetents (dangerous I know) and given that everybody lies, I think there is a distinct possibility that the survey is really about something other than its cover story.


By CaptainBlack (not verified) on 19 Oct 2010 #permalink