Newspapers as Historical Sources

          When I started to read Franzosi's article about data collections from newspapers it was going slightly over my head. I didn't think that the whole discussion about theoretical and empirical concepts was of great importance. However, the points that he puts across in the paper are very valid and actually important to anyone studying for a masters or doing a type of research for that matter.

          He says that newspapers are an imformal source of politicial and historical data but that their vailidity and reliabilty are questionable. As soon as I read that bit I started to tune in a bit more. For my dissertation last year on the history and effect of gang violence in Limerick, I relied heavily on newspaper articles as sources for my research. He says often there is no other source available, and I definately found that to be true last year, there hadn't been any real academic research done in that area before yet it was always a topic of discussion in newspapers so that was where I turned. As I intend to develop this topic further for my thesis, I know I'll be faced with the same problem, so the lesson on the validity of newspapers turns out to be an important one.

          Of course everyone knows that you can't always believe what you read in the papers, particularly the tabloids that often print out and out lies. The bigger problem is one that affects all publications, and it's a problem that is often overlooked. "Validity" doesn't question "what" the newspapers are printing but rather "how" they are printing it. That's the best way I can think of explaining it with out a long winded repitition of the entire article. The problem of editorial bias in newspapers or in books, journals and even television documentaries, is something which we as researchers can never avoid. Ten different newspapers can run the same story with the same facts but can give very different perspectives to the reader. How can we tell what a journalist was thinking 100 years after he or she writes an aricle. Impossible.

          The second problem he addressed was the percentage of coverage that certain types of stories receive in relation to their percentage of occurance. I understood this perfectly, I remember trying to prove that the media blows the level of gang violence in Limerick city way out of proportion. I found figures in a book on crime in the media that proved my point, not just about Limerick City but the whole topic. For example, murder made up 12.3% of newspaper reports at a given time, but only made up .004% of Garda reports at the same time. I think that's essentially the point he's trying to make about using a control to show the validity of articles being printed. In my example the Garda Report was the control.

          The explanations he gave about coding seemed a bit drawn out but were esay enough to understand and agree with. It's all about avoiding human error. All this aside, he doesn't think that newpapers should be completely dismissed as research sources and I agree. Errors can be found everwhere, so we can't dismiss everything.