Presentation on theme: "Doing a Dissertation Step 1: Choosing your subject Choose something that is: original uses data that have not been used before manageable data exist project."— Presentation transcript:
Doing a Dissertation Step 1: Choosing your subject Choose something that is: original uses data that have not been used before manageable data exist project is not too ambitious you have the competence interestingto you
Step 2: Planning Draw up a timetable that realistically states when you expect to have achieved each stage of the process
Step 3: Getting started Look for: ideas in the literature that you read for this and other courses literature by using book lists / course sheets references at end of articles that you read the journals in the library (browse through T6 and current serials on the ground floor) Web of Science to search for published articles on your topic (get access details for information desk in library)Web of Science NetEc to search for recent and hitherto unpublished working papers on your topicNetEc data – see lecture on datasets
Step 4: Literature review A typical dissertation may make reference to 30-50 papers or books Aim to provide an integrative and critical literature review – not an annotated list Cite references in the standard way, at the end of the dissertation – eg in the text: Freeman (1995) and Wood (1994) have argued that trade has adversely impacted those on the bottom end of the labour market. Freeman, Richard B. (1995) Are your wages set in Beijing?, Journal of Economic Perspectives, 9(3), 15-32. Wood, Adrian (1994) North-South Trade, Employment and Inequality, Oxford: Clarendon Press.
Step 5: Conceptual chapter Draw in ideas from economic theory, and show how your work will test these Does not need to be very long, eg 3-6 pages Step 6: Description of dataset Statement of what data you are using and where you got them from How many observations? Over what period? What country/ies? Provide a table of means and standard deviations for all the main variables in your analysis, and briefly discuss these. If your data are time series, discuss trends.
Step 7: Data analysis Use Excel or another package to perform analyses using: graphical methods comparison of summary statistics (eg means) over time or over subgroups regression
Example: analysis of the determinants of consumer expenditure in UK, 1950-1998 (i) Graphical analysis – consumption v income
(iii) More refined regression analysis – change in consumption depends on change in income, lagged consumption and lagged income (an error correction mechanism)
In addition to reporting the results of your analysis, you need to provide a commentary which says why youve done what youve done, and what the results imply, and which ties it in to the literature eg - the consumption function example suggests that the short run marginal propensity to consume is 0.62 and the long run marginal propensity to consume is 0.82, and this accords with the permanent income hypothesis of consumption Wheres the beef? Its in the data analysis section. The most challenging and important part of your dissertation – and also the most interesting
Step 8: Write the conclusion and introduction … … at the very end, when you know what you have discovered