Thread by @DToshkov

Did you ever want to get government positions from party-level data?
You're in luck!
Here is how to use some #rstats functions that calculate party positions & salience from @manifesto_proj data & aggregate these into gov. positions & over time @ParlGov bit.ly/2DMPLuh

for the functions to work, you need to get Manifesto Corpus (@manifesto_proj ) & @ParlGov data and follow the tutorial to align these two data sources. It was no fun identifying misaligned election dates and party codes, but now YOU don't have to do this.
Then, it's a breeze.

You can create your own scales for positions *and* for for salience from Manifesto items using different scaling approaches based on statements' shares, numbers, Lowe et al. 2011, Kim & Fording 2002.
You can aggregate these per government, per month, and per year.

And you get simple government-level average, weighted average per party seat share, min, max, range, n of parties and share missing data.
You want something else: let me know or fork the project on GitHub github.com/demetriodor/go…

Enjoy and use responsibly! Not everything that can be calculated should be calculated😃
and if you notice any mistakes or you want to streamline or extend the code, let me know or go directly to the GitHub repository linked above