I am planning a project involving critical discourse analysis (CDA) right now. How much working time should I estimate for a given amount of data? Is there any criteria for a qualified estimation?
Does this help you -
https://www.academia.edu/3800430/Introduction_to_Critical_Discourse_Analysis_2010_
Thank you so much. However, I think the paper is more about theory than practical issues of project planning. Anyway I seemed quite an inspiring read to me and I will add reading Fairclough to my bucket list. Thanks for the hint.
If you are handling data collection and general pre-processing yourself, chances are you'll be quite familiar with the material, even before sitting down to do the actual analysis. Having your protocol down (eg. annotations schemes, category assignment decision trees, etc) in advance also shaves off time from the analysis itself. Unless you are going in for a preleminary exploratory analysis before even formulating a hypothesis (or guiding research principle), the analysis should pan out to be the minor third of your project ( the other two thirds being collecting data and figuring out exact methodology).
Thank you so much, Asterios Chardalias . We already did an exploratory research and are already quite sure in which direction it will go, i.e, have some hypotheses about our field. Regarding categories, I think this will still change and develop as we follow a more inductive approach. But as our exploratory phase was about three weeks full time, I guess we can handle the rest and come to presentable results within one to two month of work.
Thank you for your answer.
Switching around categories as you go along means you'd then have to backtrack over material you have already reviewed to make sure it's up to date with your current scheme. This will add time for the analysis and a lot of frustration for work already done being thrown out the window.
It also adds confusion over which vesrion is actually the latest (thus the currently implementable) and messes with the flow for your annotation throughput. If there are more annotators than just you working on this, you'd need a robust revision control system in place and serious communication to keep everybody 'on the same page' (more time spend there).
Inductive may be great and all theory-wise, but practically it means you do things over and over. It's not the best route to go if you are on a tight schedule to 'presentable results'. I'm not saying you stick with a scheme no matter what (though time-wise that would be sweet). I'm saying chose your scheme wisely and keep alterations to a minimum. Ask yourself if the 'expressiveness'/'analyticity' gained from any alteration is actually worth your time and labor in 'impact'.
You may also just jot down any ideas for different schemes, but put them aside for now and then come back to work on your data again using those for a seperate paper. Oh, and a 'strategically' used 'Other'/'Unsure' category/tag could get you out of pinch many a times. It doesn't look the best for your stats, but it could work great if you have a narrow enough focus.
Best of luck with your project.
- Badges
- Science topic
- Similar topics
- Thesis Research