This week we looked at Roberto Franzosi's piece on using newspapers as historical sources. I found this very interesting as one could easily and often forget that newspapers are not always accurate. When using newspapers as historical sources it is important to remember this, that while they are useful their validity and reliability can be questionable. Measurement problems that go hand in hand with primary data collection from newspapers are only 'received token if any consideration' which results with a neglecting of the ways in which the measurement process may affect the results of statistical analyses as well as our theoretical conclusions. The paper discusses issues of measurement errors as well as selecting which newspapers to select to use as sources and offers solutions to problems associated with the inaccuracy of newspapers. The writer explains that the main recommendations of the paper is that more emphasis should be placed on problems of validity rather than reliability.
Validity: 'an instrument which measures exactly what it is supposed to measure and nothing else'. Empirical component: 'operationalizing and giving empirical representation to abstract concepts'. Reliability: 'concerns the degree of respectability and consistency of empirical measurements'. Such terms are discussed by Franzosi as vital to consider when using the informal sources of historical data that is newspapers. He informs us that newspapers have been most relied on as sources of data for the study of collective actions and political violence. When using them for this purpose there is always the problem of the source being questionable whether it is valid or not. He states that due to this we are left with the thought of wheater 'we are analyzing patterns of historical events or patterns of news reporting' as newspapers are often biased and selective. For example I read recently about the media coverage and in particular newspaper coverage of the 1916 Rising. James Connolly's death was reported to have happened on four different dates at four different places!! This shows how newspapers are often unreliable.
Though they are limited sometimes they are the only source available to historians to research a particular part of history as there may not be any other surviving or available source to interpret. He importantly notes that of course newspapers are not the only sources with error as many sources are rarely without some kind of error, different papers too vary as some facts will be in all reports but descriptions and other points of information could be varied. He also interestingly notes that the bias is more of silence and emphasis rather than false information. There are several things to remember when using newspapers in respect to the possibility of them being biased, such as geographical bias as well as the papers ideological stance and of course the journalist own personal opinions and views on things. Solutions are offered in how to avoid problems, such a helpful solution which is suggested is; to compare papers with other papers, as much as possible from different geographical locations and different ideological stances. By doing so similarities will show up which can thus then be taken as accurate. While dates and precise times are often very important in historical research newspapers rarely offer an exact date and time of events.
'Are newspapers data valid? My answer is, generally no'. They are likely he says to suffer from stematic error and are so not valid. 'data collection from newspapers are reliable. My answer is, generally yes'. Therefore Franzosi comes to the conclusion that papers may be reliable but may not be valid, therefore it is important to keep this in mind as they can be often the only source available for research, which shows their importance yet can only be used effectively by remembering the limitations they behold. 'Validation of data through alternative and independent measures should be an integral part of any research design involving data collection from newspapers'. This final quote sums up Franzosi's view in this paper and the message perhaps he was aiming to put across. It is a very valid one which I believe would be useful for people using newspapers as sources for their thesis topic in particular because it is important to try and get the most reliable information available obviously, to document the historical area one is interested in.
No comments:
Post a Comment