The Manual Revision of the TERENCE Italian Smart Games - 2nd International Workshop on Evidence-Based Technology Enhanced Learning - page 31

for each typology, it was necessary to work for circa 76 minutes. By considering the
need for reading the related story, filling the excel, the average time of each operator
to finalise a set of games was equal to circa 90 minutes, i.e. approx 15 minutes per
game (see Tab. 3).

Ta b l e 3 Details about the revised games

GAME n % Average time

WHO 25 10.00 10,6

WHAT 34 13.60 12

BEFORE/AFTER 74 29.60 12,8

BEFORE/WHILE 41 16.40 12,8

WHILE/AFTER 42 16.80 12,8

BEFORE/WHILE/AFTER 34 13.60 14,8

Total 250 100.00 75,8

Only in the 6% of all cases, it was necessary to change the automatically gen-
erated fixed event. In 72% of games the text of the event was corrected. The total
number of changes (of both entities or choice events) was 120. The changes were
necessary especially for the wrong choices, with 54 total changes (Fig. 4).

Fig. 4 Distribution of changes

The work of developing manual causal games was instead longer, as the operator
had to invent the game by directly using, when possible, one of the fixed events
already present in the story games. Overall, 75 causal games were created (both
cause, effect, and cause/effect games). The average time spent for their development
was equal to 23 minutes per game. The total work, also including game loading and
final review, was about 30 minutes for each causal game.