Within group - exposing participants to all conditions more than once. Two reasons why would you do that - pull out as much data as you can. Presenting iv many times and comparing performance on dv, see if there are big changes. More practical method is block randomization i. ii. iii. Between-groups: create block and that"s the order different participants participate in experiment. In within groups: it"s the same thing but each block involves all levels of independent variable in some random order, and for each block you create for a participant, you re-randomize again and again. Less practical method is abba counterbalancing (reverse counterbalancing) i. ii. iii. iv. Present levels of iv in mirror order: create random order of levels of. Iv, present them to participant forwards and then backwards. You want no level to have an advantage position-wise. Average out the differences that might come from being in a certain position.