|
Previous week Other date About iLind.net Search Contact us |
|||||
This thoughtful comment on surveys came in yesterday from a former Hawaii journalist gone academic:Your recent discussion about survey validity prompts me to respond. I know nothing of the methodology of the Star-Bulletin reader survey, but there is nothing inherently wrong in surveying specific groups, such as Star-Bulletin readers, if your research question specifically pertains to the opinions of that group. It works the other way, too -- you wouldn't ask readers a series of questions to get at why they don't read the paper.What IS important to ensure valid results is to:
(1) make sure that respondents are randomly selected (i.e. phoned at random, stopped at the mall, etc.). If respondents are self-selected (such as filling out and send in a survey printed in the paper, or calling into a TV phone line), there is a great possibility that you will miss large chunks of opinion out there. If you think of the range of opinion as a bell curve, the ones most likely to get off their butts to send in a survey are those who either are very adoring, or very angry.
And (2), the wording and sequence of questions, and the range allowed for answers, should be carefully considered to guard against "forced" responses. As one of your correspondents noted, questions that allow only yes or no type answers are likely to result in data without depth and range, and thus fairly worthless for learning -- which should be the point for any research.
The last word, perhaps?
"Get a life!"That seems to be the standard preface these days to messages objecting to my criticisms or commentaries. I suppose the idea is that if I've got enough time to offer up criticism, it must mean that I just don't have enough to do and should get back on the daily treadmill so that there won't be time to worry about such things. But I don't take it personally, and most of the reactions are actually made in good spirits.
Yesterday's reaction was from Bob Loy, news director at KGMB:
i hope people are smarter than you think. i believe most viewers understand the term "psychic." they're able to form their own opinions about whether this poor family's efforts are likely to bear fruit, or if they're being taken advantage of. there's no need for the media to explain the "psychic's "qualifications--or lack thereof.perhaps you should be explaining the evidence proving superstition, ignorance, and antiscientific views are increasing their foothold. that's an interesting statement.
It's cold this morning and a bit wet outside, with heavy rain and thunderstorms predicted for the weekend, the kind of weather that puts a crimp in the wanderings of our cats and forces them to interact in closer quarters.
I'm also paying the morning-after price for last night's experimental stop at a Thai restaurant in Kaneohe (Chao Phya in the Windward City Shopping Center), which we've driven by for years. We finally turned in as we drove past on the way home last night. This could change our routines. The food was very good. We asked for hot, and could have been hotter, never quite reaching the "I can't touch my tongue and please pass the Kleenex" level, but we can fine tune such things. I would rate it highly, especially in light of its convenient location. Well, convenient for us, at least. There just aren't many decent restaurants in Kaneohe, and almost no good ones. This little place sneaks into our "good" category, I believe.
Did you happen to catch the 6 o'clock news on KGMB (the local CBS affiliate) on Tuesday night? They ran a story about the continuing search for a college student who disappeared on Maui last month, apparently after taking LSD and becoming disoriented. Sounds ok, but this particular story featured the "psychic" hired by the family to lead this phase of the search. The report hung breathlessly on the findings of the "psychic", even displaying a drawing of a house she "saw" that might be involved. Nowhere a hint that "psychic" is a category any different from DNA chemist, pathologist, or other expert. No hints that this is fringe, nowhere-else-to-turn, potentially fraudulent stuff. Cover the "psychic", but at least be sure everybody understands the score. I hate to be a spoil sport, but this lack of critical reporting is perhaps one reason superstition, ignorance, and antiscientific views are increasing their foothold.Lucky you live Hawaii, though. Yesterday's Star-Bulletin reported the latest "most wanted" list from the U.S. Marshals Service. There are the usual bank robbers, and a drug dealer with an arsenal of weapons and explosives. But perhaps there weren't enough of those, so right up we've also got the guy convicted of conspiracy to sell illegal cable descramblers, and the crooked tax preparer who submitted tax returns claiming fake stock losses. Crooks? Sure. Most wanted desperados? As I said, lucky you live here.
Seems I stepped on some Star-Bulletin toes with Monday's report on fragments of their readers poll results. "We were asked not to let it out of the newsroom just yet," according to one newsroom staffer. But didn't I see house ads in the Bulletin over the last couple of weeks inviting people to sign up for a presentation of the latest survey results? That doesn't sound much like "top secret".
Then this request: "Do you have any hard (or honest) figures from the Advertiser's surveys? If so, make 'em public and level the playing field. Be fair."
I'm told that the Advertiser's survey results have not been discussed in their newsroom. Over there, it's apparently management only, confidential stuff.
A couple of comments came in yesterday on the question of newspaper research."The validity of the Star-Bulletin's research is greatly diminished because its sample came from existing readers," according to one. In this view, apparently, the only survey worth doing is aimed at the universe of potential readers, instead of actual readers. But it seems to me that both offer substantial and robust groups that are necessary to understand. You would obviously ask different questions to each group, and glean different sorts of insights, but each appears to be equally valid, at least as long as the group of actual readers isn't later described as if it comprised that much larger group of potential readers. Is that a dangerous temptation in the current newspaper battle? I don't know.
Another commented on the pitfalls of polling:
We used to run surveys in our sports section and I recall having to call some of the respondents to ask them why they had said "yes" or "no" to a question. What I was annoyed to find out was that sometimes the respondents had no idea why they answered one way or the other because the surveyors had thrown questions about three or four questions at them at the same time. I was told by some of these respondents that they might have had enough understanding of one, maybe two questions, to answer, but not enough about the others to answer in a knowledgeable way. They said they were pressured into answering all of them anyway.It was a pleasure watching the segment from Boston on last night's News Hour on PBS, which featured the editor and investigative team leader at the Boston Globe discussing their blockbuster series on sexual abuse by priests. Not because of the depressing subject, but because you can always learn from a great project. I was especially interested in their creative use of data. When denied official data, they gathered years of old church directories, painstakingly entered data on where priests were assigned, and then tracked unusual or repeated movements, which they guessed might be a signal of other problems. Further investigation of these cases led them to some of the "repeat offenders".
Pacific Business News reports in this week's issue that the union representing employees in the Advertiser's press room has filed to trigger a union election at MidWeek's Kaneohe printing plant. The Graphic Communications International Union gathered union authorization cards from MidWeek employees beginning last summer, and filed their petition with the NLRB in December. The move is being strongly resisted by the management of the Star-Bulletin/MidWeek, which has thrown considerable legal firepower into delaying or defeating the organizing drive. A ruling by the NLRB defining the prospective bargaining unit is pending.A question from a reader yesterday questioned the validity of the Star-Bulletin readers' poll because respondents were not selected randomly and could not, statistically speaking, be used to predict the views of all Star-Bulletin readers. It's certainly true that survey analysis presumes a random sample of the larger population. It isn't known whether this S-B poll tried to correct its numbers by adjusting for differences between known S-B readership and the demographics of those who responded to this survey.
On the other hand, it's certainly a lot less expensive to print a survey in your own newspaper and take whatever responses are received than to hire a survey team to generate a real random sample of readers. Under the circumstances, though, some data at reasonable cost is probably preferable to ideal data at a prohibitive cost.
Phil Currie, Gannett's Senior Vice President for News, has responded to criticisms in a recent book by two Washington Post editors, His comments, which appear on the Gannett web site, make worthwhile reading.
Perhaps I just haven't been observant enough recently, but I was surprised yesterday to see direct in-your-face competition in Sunday newspaper street sales at a number of major intersections. In many cases, two hawkers would walk along adjacent lanes, one selling Star-Bulletin's and the other selling Advertisers. I was more used to seeing them staking out nearby corners instead of challenging on the same turf. It's a war out there.Star-Bulletin newsroom staffers got a glimpse of the recent readers poll results on Friday. Overall, on a scale of 5 (excellent) down to 1 (very poor), readers gave the S-B a 3.74 rating. The survey firm, Pulse Research, said most papers fall between 3.25 and 3.50 overall, so viewed this finding as positive.
When asked to identify the three most important issues, Star-Bulletin readers said:
Economy 66.2%
Education 62.3%
State and local government 54.5%
Crime 45.3 %
Environment 29.8%
Traffic 18%
Other 4%Interesting results, although such issue ranks are notoriously fickle and reflect the existing news emphasis as much or more than independent reader judgments.
One Star-Bulletin newsroom staffer had this reaction to the survey:
Our staff really needed the boost in morale. The way I see it, it vindicates our approach to newspapering. Yeah, the old SB is gone, but given the parameters under which we operate, we're doing pretty good with the readers. It doesn't look like we will be forced to publish another 'Mudweak' anytime soon.I try, believe it or not, to try to take all of this with a grain of salt. After all, given the adage that liars can figure and figures can lie, if you phrased the questions of a survey the right way you can come up with the results that the moon is made of green cheese and there is a man in it.
There was a price war among competing street hawkers around noon yesterday. It was in the shadow of the freeway on Lunalilo Street, between Piikoi and Pensacola. As cars rounded the corner to head towards the freeway entrance, they were greeted first by someone selling the Advertiser. Just down the block, and enthusiastic Star-Bulletin hawker displayed his wares and a sign, crudely handwritten on a piece of cardboard, indicating a sale price of 35 cents.I was taken to task yesterday by someone writing as "Leo Silverman":
"In the news war, now there are rumblings that Gannett wants to follow the Star-Bulletin's lead with its own round of layoffs at the Advertiser."Rumblings? In the computer industry, we call that FUD - fear, uncertainty and doubt. The Nixon administration, as you're undoubtedly aware, would have called that kind of rumormongering ratfucking. Seems pretty weak. I'm still waiting for the announcement of those Advertiser pay cuts the rumblers were predicting. Or the death of the Advertiser's afternoon edition that the rumblers twice foretold. It seems like this rumbling is nothing but the wishful thinking of bitter Bulletin folks, or disenfranchised Advertiser folks. Unless you actually have sources privy to Gannett Corporate's thinking, it would seem you're letting yourself get played. I thought you had more self-respect than that.
I don't see it this way. The rumblings may reflect wishful thinking on the part of some. But they're also part of the mood there on South Street. I don't present such things as true indicators of corporate decisions to come, but as reflections of what's being talked about among the troops. And with the Guild contract at the Advertiser expiring in mid-June, bargaining is underway, providing a glimpse of Gannett's preferred future and a natural source of rumors and hallway chatter.
Neighbor Bob sent along a New York Times story by Katharine Q. Seelye reprinted in Friday's International Herald Tribune and an accompanying photo.
Bob comments:
if you see the picture first
then read the article
is not all of this from a novel we read in high school
i think it was called "1984"
subtitle: it can't happen here
subtitle: this happens in communist countries only
lord have mercy
then again ... why lob grenades into a church
a whole lot'a hate go'in on here
And he drew this thoughtful reply from another friend:Funny the author should mention those guys (Hitler and Stalin) in the end. Have any of you checked out history about the German Catholic Movement, which swept Europe, and even parts of the US, in the 30s? An oncoming apocalyptic battle between Good and Evil, won after long years by Good, after which a thousand-year-reign for you-know-who, under which peace progress and prosperity would prevail. The Vatican bought into it. Once the evil races, and the people with poisonous thought processes were eradicated, all would be for the greater glory of God, and his swift sword, the military of the most righteous and modern nation. (Robert McAfee Brown, "On saying yes and saying no".)We skeptics should look at some of the mythology that underpins this seemingly mindless support for our new crusade against evil.
Provocative thoughts for a Sunday morning.
Search this site,
courtesy of the folks at Atomz.com
\*/.