The Political Party scale was determined by selecting a set of questions from Pew Research Center’s March 2016 political survey representing a range of policy issues and political values that are each consistently associated with party identification. A logistic regression, a technique employed to estimate the independent contribution of multiple factors in predicting a particular outcome, was used to predict the likelihood of being Republican or Democrat. The result of the logistic regression analysis is a set of coefficients, or relative weights, for each question in the quiz. Therefore, each question has a different impact on the overall score, depending on how strongly each item is related to partisanship. When you take the quiz, your answers are weighted by these coefficients to produce your overall score: The answer to each question is multiplied by the coefficient for that question, the products are summed up (along with a constant term) and you are placed on the scale from most liberal to most conservative.
Because the theoretical range of the scale is much wider than the actual distribution of most of the public, the tails of the scale (scores beyond the value of the “most ideological” possibility at both ends of the scale) are recoded to the highest and lowest possible scores respectively on the scale. This compression makes it easier to see the differences between groups in the graphs. The generated result will tell you which group you are closest to on the scale.
The center of the scale is the score of the average respondent, which is the mean score for all Americans. To determine the placement of each of the partisan and demographic groups shown on the scale, the mean of the scores for all members of each group was calculated.