The functioning of central categories Middle Level and Sometimes in graded response scales: does the label matter?

The present study evaluates the extent to which central categories explicitly labeled as being in the middle of the other response categories, specifically Middle Level and Sometimes, function as expected according to the integer scoring system. The assumptions are tested by means of Bock's Nominal Model in two 5-response scales. Results show that the assumption of the ordering of the response categories is met for all the items. The ordering of thresholds is satisfied for all but one item with the central category Middle Level . Results are compared with those obtained when middle categories are not explicitly labeled as being in the middle of the other response categories, as in the case of Not Sure , Undecided or?