The Chronicle ran a Gebe Martinez story with the following headline yesterday:
Poll ties economics, school size to higher Latino dropout rate
The Chronicle‘s headline writing is frequently weak, but this one is horrible even by Chron standards, since the story is about a Pew study that analyzes characteristics of high schools attended by Hispanics. The story uses the term “survey,” but it’s not referring to polling. It’s referring to a study, which should have been clear from the context of the story. The study made use of US Department of Education data, not polling data.
Furthermore, both the headline and the story contain overly broad conclusions. The Pew study is important because it suggests some unique characteristics of schools attended predominantly by Hispanics. One could use that information to construct a properly specified social scientific model in order to test the impact of school characteristics (and other independent variables) on educational outcomes. However, this Pew study did not actually do that (although the principal investigator, Richard Fry, does suggest that the different characteristics might explain differing educational outcomes, he makes no such claim that this study actually demonstrates that linkage; it does not), and both the headline and the substance of the story are erroneous to the extent they suggest it did.
It’s possible that neither the headline writer nor Martinez has special expertise in social scientific modelling, which might explain the overly broad characterization.
The press release, executive summary, and full report are all available on the Pew Hispanic Center site. It might have been useful for Chron.com to provide links, since some readers actually do prefer to read source materials for themselves (which, in this case, is necessary for a more accurate understanding of the study).
