Group items tagged

Journalists often make mistakes when reporting on data such as opinion poll results, federal jobs reports and census surveys because they don't quite understand - or they ignore - the data's margin of error. Data collected from a sample of the population will never perfectly represent the population as a whole. The margin of error, which depends primarily on sample size, is a measure of how precise the estimate is. The margin of error for an opinion poll indicates how close the match is likely to be between the responses of the people in the poll and those of the population as a whole. To help journalists understand margin of error and how to correctly interpret data from polls and surveys, we've put together a list of seven tips, Look for the margin of error - and report it. It tells you and your audience how much the results can vary. Remember that the larger the margin of error, the greater the likelihood the survey estimate will be inaccurate. Make sure a political candidate really has the lead before you report it. Note that there are real trends, and then there are mistaken claims of a trend. Watch your adjectives. (And it might be best to avoid them altogether.) Keep in mind that the margin of error for subgroups of a sample will always be larger than the margin of error for the sample. Use caution when comparing results from different polls and surveys, especially those conducted by different organizations.

The plan to use the Australian Bureau of Statistics to conduct the federal government's postal plebiscite on marriage reform raises an interesting question: wouldn't it be easier, and just as accurate, to ask the ABS to poll a representative sample of the Australian population rather than everyone?

On the importance of having data, in order to understand and tackle an issue... The most recent U.S. estimate...concluded approximately 513,000 women and girls at risk of genital mutilation...But that number should be taken with a big grain of salt...the data doesn't account for immigrants from countries where female genital cutting isn't studied or widely practiced...."You also can't assume that people who come to the U.S. are a representative sample of their country of origin," Clark said. That's especially problematic for estimating rates of female genital cutting, since it's not practiced uniformly within countries. It's also possible, he said, that some immigrants abandon the procedure as they assimilate....some advocates point out that although the estimates focus on immigrants, ...female genital cutting isn't new to the U.S. Female circumcision was performed as a treatment for masturbation by American physicians as recently as the mid-20th century...

Say you have a 3,000-person presidential election survey from a state where 3 percent of the population is black. If your survey is exactly representative of reality, you'd end up with 90 black people out of that 3,000. Then you ask them who they plan to vote for (for our purposes, we're assuming they're all voting). History suggests the vast majority will go with the Democrat. Over the last five presidential elections, Republicans have earned an average of only 7 percent of the black vote nationwide. However, your survey comes back with 19.5 percent of black voters leaning Republican. Now, that's the sort of unexpected result that's likely to draw the attention of a social scientist (or a curious journalist). But it should also make them suspicious. That's because when you're focusing on a tiny population like the black voters of a state with few black citizens, even a measurement error rate of 1 percent can produce an outcome that's wildly different from reality. That error could come from white voters who clicked the wrong box and misidentified their race. It could come from black voters who meant to say they were voting Democratic. In any event, the combination of an imbalanced sample ratio and measurement error can be deadly to attempts at deriving meaning from numbers - a grand piano dangling from a rope above a crenulated, four-tiered wedding cake. Just a handful of miscategorized people and - crash! - your beautiful, fascinating insight collapses into a messy disaster.

Should newspapers be adding confidence intervals to their graphics? Why, she asked, are newspapers like hers hesitant to print confidence intervals, a statistical measure of uncertainty? With the exception of noting sampling error in polling data, newspapers like the Times only show uncertainty when they're forced to - and often to prove the opposite of what point data might show.

Data aren't dead, contrary to what some pundits stated post-election [2], rather the limitations of data are not always well reported. While pollsters will be reworking their models following the election, what can media journalists do to improve their overall coverage of statistical issues in the future? First, discuss possible statistical biases, such as errors in sampling and polling, and what impact these might have on the results. Second, always provide measures of uncertainty, and root these uncertainties in real-world examples.