For certain lists, however, randomization would introduce confusion, and would therefore not be appropriate. Generally these are items respondents are accustomed to seeing in a certain order:
This is an example of randomization in a multi-select question. These options are not directly related to each other, and thus they appear in random order. However, you’ll notice that “Other” is fixed at the bottom of this list, which means that it always appears as the last option. This is done because “Other” is used to capture options not the in list, and placing it in a different spot would be confusing for respondents. Thus, it needs to be anchored at the bottom.
This is an example of random reversal in a single-select question. Since this is a linear scale, randomization would be confusing. Instead, these options are shown only in this order (negative to positive), or in reverse order (positive to negative).
This is an example of question randomization. These two questions have the same option lists, and the question structure is similar. If there were three or more of such questions in a row, randomizing their order could help control for the effects of survey fatigue.