The effect of adaptivity on the reliability coefficient in adaptive comparative judgement (Bramley & Vitello, 2019)

Data from a Comparative Judgement assessment exercise of 150 GCSE English scripts.There are data from three studies (labelling as in Bramley & Vitello (2019)): 1a. data produced where comparisons are scheduled adaptively; 1b. data produced from comparing (nearly) all of a subset of 20 items in a round-robin format; 2. data produced where comparisons are scheduled randomly as well as repeated comparisons of some particular pairs.The data for each study consists of four columns: Judge - an identifying number for the judge (note that the judges in 1a and 1b are the same, but different to those in 2); Item 1 - identifier for first script in the comparison; Item 2 - identifier for second script in the comparison; Win - the result Win=1 / Loss =0 from perspective of Item 1. Note that each pairwise comparison is shown as two rows. This is an artefact of the system used to originally collect the data. The second row for each comparison is redundant.Full details of the data and its collection can be found in "The effect of adaptivity on the reliability coefficient in adaptive comparative judgement" (Bramley & Vitello, 2019) https://doi.org/10.1080/0969594X.2017.1418734