Skip to main content
. 2018 Nov 3;37(2):201–226. doi: 10.1007/s40273-018-0734-2

Table 3.

Experimental design information DCEs

Item Category 1990–2000 2001–2008 2009–2012 Current: 2013–2017
N = 34a (%)b N = 114a (%)b N = 179a (%)b N = 301a (%)b
Design type Full factorial 4 (12) 0 (0) 9 (5) 13 (4)
Fractional 25 (74) 114 (100) 158 (88) 269 (89)
Not clearly reported 5 (15) 0 (0) 12 (7) 19 (6)
Design plan Main effects only 25 (74) 100 (89) 98 (55) 86 (29)
Main effects and two-way interactions 2 (6) 6 (5) 23 (13) 52 (17)
Not applicable 4 (12) 0 (0) 5 (3) 5 (2)
Other N/C N/C N/C N/C N/C N/C 11 (4)
Not clearly reported 3 (9) 8 (7) 52 (29) 147 (49)
Blocking Yes N/C N/C N/C N/C N/C N/C 150 (50)
No N/C N/C N/C N/C N/C N/C 60 (20)
Not clearly reported N/C N/C N/C N/C N/C N/C 91 (30)
Design software Ngene N/C N/C N/C N/C N/C N/C 62 (21)
SAS 0 (0) 14 (12) 41 (23) 54 (18)
Sawtooth 2 (6) 5 (4) 30 (17) 47 (16)
SPEED 13 (38) 22 (19) 9 (5) 1 (0)
SPSS 2 (6) 14 (12) 13 (7) 20 (7)
Not applicable N/C N/C N/C N/C N/C N/C 11 (3)
Other 2 (6) N/C N/C 27 (15) 7 (2)
Not clearly reported N/C N/C 4 (4) 9 (5) 99 (33)
Design source Website 0 (0) 3 (3) 9 (5) 4 (1)
Expert 4 (12) 4 (4) 11 (6) 5 (2)
Not clearly reported 9 (26) 42 (37) 30 (17) 215 (71)
Methods to create choice sets Orthogonal: single profiles (binary choices) 3 (9) 12 (11) 2 (1) 7 (2)
Orthogonal: random pairing 18 (53) 19 (17) 18 (10) 12 (4)
Orthogonal: pairing with constant comparator 6 (18) 23 (20) 5 (3) 0 (0)
Orthogonal: foldover-random pairing 0 (0) 1 (1) 4 (2) 2 (1)
Orthogonal: foldover 0 (0) 11 (10) 34 (19) 26 (9)
D-efficiency 0 (0) 14 (12) 54 (30) 105 (35)
Bayesian D-efficiency N/C N/C N/C N/C N/C N/C 23 (8)
Other 4 (12) 2 (2) 27 (15) 26 (9)
Not clearly reported 3 (9) 32 (28) 39 (22) 100 (33)

DCE discrete choice experiment, N/C not collected (data were not collected for this specific category)

aNumbers of individual studies might not add up to total Ns as some studies addressed multiple topics

bPercentages might not add up to 100% because some studies addressed multiple topics and because of rounding error