#Sample Jupyter notebook for analysis of open questions
##The anonymised data is drawn from a short survey of teaching and support staff and managers at three NZ tertiary institutions run in 2015/16
- Click on LA Questionnaire Analysis.ipynb to launch the analysis notebook
- When prompted, select Python 2 as the kernel
- Wait for the notebook to load the go to the cell menu and choose run all
- It takes a moment for the cells to execute.
- Try changing the question number (In ). See variable openqs (In ) to see list of available questions. Go to cell menu and choose ‘run all below’. This executes the current cell and everything below it.
- Try changing the myword variable to run a concordance (In ) for that word. Suggest choosing a word from the list of top 30 words in the response corpus.Again, you’ll need to go to cell menu and choose ‘run all below’ to see the changes in output.
- Scroll down In  to see the concordance visualised as a Wordtree.
##There seem to be minor issues with Google Charts in notebook. The Google charts API is used for the Wordtree visualisation:
If Wordtree is not loaded, save any changes you have made, then refresh the entire page. Go to cell menu and choose run cells. The Wordtree should display.
If Wordtree still does not load and there is an error, click in In  and choose run cells from the cell menu. This just executes that cell.
Wordtree is from Google Chart.
You can click on any word to follow that branch.
Word weight reflects word frequency relative to keyword in context.