A new metric of consensus for Likert-type scale questionnaires: an application to consumer expectations

In this study, we present a metric of consensus for Likert-type scales. The statistic provides the level agreement any given number response options as percentage among respondents. With aim, use geometric framework that allows us to analytically derive positional indicator. is obtained relative weight distance from point containing proportions observations fall in each category centre regular polygon with many vertices categories, which corresponds maximum dissent. can be regarded area encompasses all possible answering combinations. order assess performance proposed consensus, conduct an iterated forecasting experiment test whether inclusion degree households’ expectations improves out-of-sample forecast accuracy unemployment rate seven European countries and Euro Area. We find evidence households contains useful information predict rates cases. This result shows potential metrics track evolution economic variables. Finally, design simulation compare sampling distribution three- five-response alternatives, finding former higher granularity dispersion.