deletions | additions       diff --git a/Further Bias Corrections (Not implemented).tex b/Further Bias Corrections (Not implemented).tex  index 25685a3..0eadfae 100644  --- a/Further Bias Corrections (Not implemented).tex  +++ b/Further Bias Corrections (Not implemented).tex 
...
\subsection{2.2}  The main bias correction mechanism that we employed was described in section 2.1.  The multiplication of the transition matrix with the 2013 results vector can be viewed as giving different respondents different weights. If Ri respondents replied that they voter for party i in 2013, then in the normalized transition matrix, each such respondent has weight 1/Ri. When we right-multiply the matrix with the actual 2013 results vector A, each respondent ends up with the weight Ai/Ri. With this weighting scheme, the total weight of respondents that claims to have voter for party i in 2013 is the same as the actual number of voters for i in 2013. Our sample now 'agrees' with the actual 2013 election results.