Use this URL to cite or link to this record in EThOS:
Title: New methods for new data : developing techniques for analysing alternative social science data
Author: Mellon, Jonathan
Awarding Body: University of Oxford
Current Institution: University of Oxford
Date of Award: 2013
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
In social science research, methodology matters. Seemingly arcane issues of measurement are not trivial but can have profound influences on how we view the social world. This thesis seeks to fill methodological gaps that have opened up as new research techniques become available. This thesis consists of five chapters covering three projects. The first project analyses whether Google Trends data can be used as proxies for issue salience. A methodology for validating Google Trends data against survey data is developed, with searches validated in the United States, the United Kingdom and Spain. The dangers of not testing the validity and reliability of Internet derived data is discussed. The second project examines the translation equivalence of survey instruments into different languages by using corpus linguistic on Internet corpora in 10 languages. Using these corpora, dissimilarity measures are constructed, allowing clusters of meaning to be identifed quantitatively and examined qualitatively. This method is illustrated using a postmaterialism item from the World Values Survey. This item is found to suffer from important translation problems, as the translations of the word ideas have clearly nonequivalent meanings across languages. These differences predict the item's correlations with other variables within the World Values Survey. The third project looks at the role of information processing within the US State Department's foreign policy analysis using social network analysis (SNA). First, the predictors of information flow are identifed, showing diplomats focus most strongly on trade. Next, a new type of relational event model is developed to account for specifc communication features communication within this bureaucracy: a relational event conversation history actor initiated network model (RECHAIN). Using this model, we fnd the importance of trade is set by the Bureaus, with Embassies adopting this agenda against their preferences. Finally, we examine the evidence behind different mechanisms of communication agenda setting.
Supervisor: Fisher, Stephen D. Sponsor: Economic and Social Research Council
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: social network analysis ; political sociology ; social science measurement ; political behaviour