Week 10: Abstracts
Margin of Error
Why must polls be so tricky? UF should have an entire class devoted to just covering all those little nuisances of polling data. Yikes. It’s something I wouldn’t want to touch with a 10-foot pole.
Robert Niles’ article is a great reference for journalists when it comes to margins of error. I think newspapers should implement an on-going training session that re-evaluates how polls will be used in articles. It should also be constantly updated and there should be a reporter briefing around the time when presidential and primary elections are in full swing. It’s hard to remember all that 95 out of 100 stuff, so reporters should be constantly reminded.
Best case scenario: No polls would ever have to be used in articles! But, that’s not likely to happen. Make sure the poll was completed by a reputable company, the questions were not biased and didn’t make the pollster answer in a desired response and that readers are given all of the correct information regarding the margin of error and exactly what that means. A little disclaimer at the bottom just doesn’t do justice.
STATS
I really liked this Web site. It’s nice to see that there’s a non-profit organization out there dedicated to cutting through all the B.S. that some newspapers put out. Just out of curiosity, I read the article demystifying a Chicago Tribune article about how dangerous moderate drinking is. As a college student, I’ve been known to enjoy a drink or two and so do my friends. I just wanted to see what STATS uncovered. Their article was hilarious. I like how they presented the side of the Chicago Tribune but then also counteracted each statement with an equally true fact. It just goes to show that you can believe everything you read.
This is a great site that is truly doing investigative reporting. They’re looking at what’s being fed to the public and trying to correct all that information – wrong or not – that’s going in. As a journalist, I’d want to bookmark this page to reference if I’m ever working on a story that is using a lot of medical and scientific claims. Sometimes it may seem like we are reporting both sides of the story when, in fact, we aren’t.
NCPP
Going back to what I said about having an on-going training session for journalists who often deal with polls, this NCPP article is a great resource. Bookmark these questions and ask them every time you come across a poll. One of the most important questions to ask is who paid for the poll. By knowing where the source of income came from, you can make an informed decision on whether or not the poll is even worthy of reporting. It can help eliminate all kinds of biased coverage. Another important piece of information is determining just how many people were polled and where exactly they came from. Is it really a random sample of Americans? Was the poll only conducted through the phone or online? These types of questions can help you determine whether a large chunk of participants are young, old, republican, democrat, middle-class, upper-class, black or white.
