reported in the 1980s was 0.86, and for the 30 most recent datasets (26 laboratory, 4 field) was 0.85. Figure H-4 shows the relationship of A to time, distinguishing between data on comparison question tests and concealed information or guilty knowledge tests. Although there are many more comparison question than concealed information datasets, the trend appears downward for both.
In view of the low methodological quality ratings that we assigned to many of these studies, we readdressed the above points in the subset of 14 laboratory studies with internal validity scores better than 3 on a 1-5 scale (with 1 as best). These studies included 721 polygraph tests. Figures H-5, H-6, and H-7 are the counterparts within this subgroup of studies of Figures 5-1, H-3, and H-4 for the laboratory studies with the best internal validity. The situation portrayed is much the same as for the full group of laboratory studies. Outliers excluded, the negative time trend in A values from comparison question test datasets appears similar in the high internal validity subgroup, and the decline is not visible in the concealed information test datasets of which, however, only three are included. In Figure H-7, each dataset is symbolized on the plot by the salience score