3.1.4 Procedure The experiment was given as an online survey. Participants indicated that they were eligible and willing to participate in the survey by clicking a consent box underneath a consent form. Then, participants were introduced to how the experiment would work, given the instructions in section 3.1.2, and given a few practice items. After finishing the practice items, participants began with the first test item. Participants were given a screen between each of the test items so that they could take breaks if they wished to do so. For each test item, participants would give the sentence a rating from 1-7 (with 1 being the worst, and 7 being the best, either by pressing a number on the computer screen or on their keyboard).
3.2 …show more content…
Using this model for a Latin square design can be advantageous, for several reasons. There is not as much need for prior averaging as there would be when using an ANOVA analysis. Subjects (participants of the study) and stimuli items can be treated as random effects in the analysis, and noise in the data can be better controlled for (Baayen, Davidson,& Bates,