In the paper I classified the UK electoral system as a strict list system. I think I made a correct choice but someone might disagree with me on this point. Here I repeat the same analysis as in the paper but the UK observations are exluded (i.e. n=59). The results show that the classification does indeed matter. If one looks at these results the conclusion would be that multiple constituencies is a more important variable than strict partly list. Go and figure it out! However, in model 1b and 3b the list variable has the expected sign and the coeffiecient is almost statistically significant. So maybe I am correct after all?
Compulsory voting ++++++ Simultanous elections ++++++ Weekend voting ++++++ Country net payer ----- Country net receiver Share of seats + Strict party lists Multiple constituencies +++++ New member country ++
--> REGRESS;Lhs=EPELEC;Rhs=ONE, compulso, arrival, concurr, seatshar, restday...
,constitu, netpayer, netrecei$
+-----------------------------------------------------------------------+
| Ordinary least squares regression Weighting variable = none |
| Dep. var. = EPELEC Mean= 62.69152542 , S.D.= 17.67353650 |
| Model size: Observations = 59, Parameters = 10, Deg.Fr.= 49 |
| Residuals: Sum of squares= 3302.043595 , Std.Dev.= 8.20906 |
| Fit: R-squared= .817733, Adjusted R-squared = .78426 |
| Model test: F[ 9, 49] = 24.43, Prob value = .00000 |
| Diagnostic: Log-L = -202.4478, Restricted(b=0) Log-L = -252.6651 |
| LogAmemiyaPrCrt.= 4.367, Akaike Info. Crt.= 7.202 |
| Autocorrel: Durbin-Watson Statistic = 1.25715, Rho = .37142 |
+-----------------------------------------------------------------------+
+---------+--------------+----------------+--------+---------+----------+
|Variable | Coefficient | Standard Error |t-ratio |P[|T|>t] | Mean of X|
+---------+--------------+----------------+--------+---------+----------+
Constant 46.39398369 2.9655269 15.644 .0000
COMPULSO 24.90475894 3.2969194 7.554 .0000 .30508475
ARRIVAL 5.359927127 4.1744657 1.284 .2052 .10169492
CONCURR 10.58629589 2.8035311 3.776 .0004 .28813559
SEATSHAR .1898178410 .37033913 .513 .6106 7.5910619
RESTDAY 3.217765529 3.9109664 .823 .4146 .74576271
LIST 4.697905187 6.5323645 .719 .4754 .38983051
CONSTITU 10.55133549 4.7439901 2.224 .0308 .25423729
NETPAYER -4.571962576 3.0327179 -1.508 .1381 .20338983
NETRECEI -7.606194198 5.0320783 -1.512 .1371 .30508475
--> REGRESS;Lhs=EPELEC;Rhs=ONE, compulso, arrival, concurr, seatshar, restday...
,constitu, netpayer, netrecei, y1999,italcomp, finland, sweden, france$
+-----------------------------------------------------------------------+
| Ordinary least squares regression Weighting variable = none |
| Dep. var. = EPELEC Mean= 62.69152542 , S.D.= 17.67353650 |
| Model size: Observations = 59, Parameters = 15, Deg.Fr.= 44 |
| Residuals: Sum of squares= 1289.852577 , Std.Dev.= 5.41432 |
| Fit: R-squared= .928802, Adjusted R-squared = .90615 |
| Model test: F[ 14, 44] = 41.00, Prob value = .00000 |
| Diagnostic: Log-L = -174.7174, Restricted(b=0) Log-L = -252.6651 |
| LogAmemiyaPrCrt.= 3.605, Akaike Info. Crt.= 6.431 |
| Autocorrel: Durbin-Watson Statistic = 1.99423, Rho = .00288 |
+-----------------------------------------------------------------------+
+---------+--------------+----------------+--------+---------+----------+
|Variable | Coefficient | Standard Error |t-ratio |P[|T|>t] | Mean of X|
+---------+--------------+----------------+--------+---------+----------+
Constant 45.76760861 2.5145439 18.201 .0000
COMPULSO 23.47711093 3.0735836 7.638 .0000 .30508475
ARRIVAL 4.361426313 3.1069790 1.404 .1674 .10169492
CONCURR 10.57509546 2.0339369 5.199 .0000 .28813559
SEATSHAR 1.030569193 .38939155 2.647 .0112 7.5910619
RESTDAY 11.94399836 3.0992107 3.854 .0004 .74576271
LIST -10.21411532 6.0363661 -1.692 .0977 .38983051
CONSTITU 3.988590709 3.5161440 1.134 .2628 .25423729
NETPAYER -8.595514518 2.6470120 -3.247 .0022 .20338983
NETRECEI -3.406436166 4.4301980 -.769 .4461 .30508475
Y1999 -6.550035928 1.8914082 -3.463 .0012 .23728814
ITALCOMP -24.17208348 5.8242248 -4.150 .0001 .50847458E-01
FINLAND -19.48889256 4.7158958 -4.133 .0002 .33898305E-01
SWEDEN -11.69359632 5.4390232 -2.150 .0371 .33898305E-01
FRANCE -10.19644500 3.7651556 -2.708 .0096 .84745763E-01
--> REJECT; country=11$
--> REGRESS;Lhs=differen;Rhs=ONE, compulso, arrival, concurr, seatshar, restd...
,constitu, netpayer, netrecei$
+-----------------------------------------------------------------------+
| Ordinary least squares regression Weighting variable = none |
| Dep. var. = DIFFEREN Mean= -17.96237288 , S.D.= 14.30231044 |
| Model size: Observations = 59, Parameters = 10, Deg.Fr.= 49 |
| Residuals: Sum of squares= 2081.088180 , Std.Dev.= 6.51699 |
| Fit: R-squared= .824592, Adjusted R-squared = .79237 |
| Model test: F[ 9, 49] = 25.59, Prob value = .00000 |
| Diagnostic: Log-L = -188.8291, Restricted(b=0) Log-L = -240.1779 |
| LogAmemiyaPrCrt.= 3.905, Akaike Info. Crt.= 6.740 |
| Autocorrel: Durbin-Watson Statistic = 1.59164, Rho = .20418 |
+-----------------------------------------------------------------------+
+---------+--------------+----------------+--------+---------+----------+
|Variable | Coefficient | Standard Error |t-ratio |P[|T|>t] | Mean of X|
+---------+--------------+----------------+--------+---------+----------+
Constant -33.88490096 2.3542669 -14.393 .0000
COMPULSO 12.40615592 2.6173522 4.740 .0000 .30508475
ARRIVAL 2.797467322 3.3140170 .844 .4027 .10169492
CONCURR 13.47510910 2.2256621 6.054 .0000 .28813559
SEATSHAR -.1015868906 .29400414 -.346 .7312 7.5910619
RESTDAY 6.584442066 3.1048307 2.121 .0390 .74576271
LIST 8.330552217 5.1859014 1.606 .1146 .38983051
CONSTITU 11.01604866 3.7661500 2.925 .0052 .25423729
NETPAYER -7.586974605 2.4076084 -3.151 .0028 .20338983
NETRECEI -2.208916937 3.9948569 -.553 .5828 .30508475
--> REJECT; country=11$
--> REGRESS;Lhs=differen;Rhs=ONE, compulso, arrival, concurr, seatshar, restd...
,constitu, netpayer, netrecei , y1999,italcomp, luxembou, sweden
$
+-----------------------------------------------------------------------+
| Ordinary least squares regression Weighting variable = none |
| Dep. var. = DIFFEREN Mean= -17.96237288 , S.D.= 14.30231044 |
| Model size: Observations = 59, Parameters = 14, Deg.Fr.= 45 |
| Residuals: Sum of squares= 1085.671849 , Std.Dev.= 4.91183 |
| Fit: R-squared= .908492, Adjusted R-squared = .88206 |
| Model test: F[ 13, 45] = 34.37, Prob value = .00000 |
| Diagnostic: Log-L = -169.6337, Restricted(b=0) Log-L = -240.1779 |
| LogAmemiyaPrCrt.= 3.396, Akaike Info. Crt.= 6.225 |
| Autocorrel: Durbin-Watson Statistic = 2.12660, Rho = -.06330 |
+-----------------------------------------------------------------------+
+---------+--------------+----------------+--------+---------+----------+
|Variable | Coefficient | Standard Error |t-ratio |P[|T|>t] | Mean of X|
+---------+--------------+----------------+--------+---------+----------+
Constant -36.41620420 2.1926655 -16.608 .0000
COMPULSO 11.71614953 2.7142542 4.317 .0001 .30508475
ARRIVAL 7.203020943 3.1085821 2.317 .0251 .10169492
CONCURR 8.459388037 2.0740786 4.079 .0002 .28813559
SEATSHAR .4479033947 .35129690 1.275 .2089 7.5910619
RESTDAY 7.671197007 2.9112560 2.635 .0115 .74576271
LIST 2.202065287 5.7091204 .386 .7015 .38983051
CONSTITU 11.78838749 3.6815768 3.202 .0025 .25423729
NETPAYER -7.455142531 2.1426027 -3.479 .0011 .20338983
NETRECEI 3.022871996 3.7185609 .813 .4205 .30508475
Y1999 -3.084742974 1.7550333 -1.758 .0856 .23728814
ITALCOMP -12.75006093 5.3506814 -2.383 .0215 .50847458E-01
LUXEMBOU 12.98014996 4.4969730 2.886 .0060 .84745763E-01
SWEDEN -11.58309065 4.7175343 -2.455 .0180 .33898305E-01
--> REGRESS;Lhs=EPELEC;Rhs=ONE, compulso, arrival, concurr, seatshar, restday...
,constitu, netpayer, netrecei, npelec$
+-----------------------------------------------------------------------+
| Ordinary least squares regression Weighting variable = none |
| Dep. var. = EPELEC Mean= 62.69152542 , S.D.= 17.67353650 |
| Model size: Observations = 59, Parameters = 11, Deg.Fr.= 48 |
| Residuals: Sum of squares= 2051.360969 , Std.Dev.= 6.53733 |
| Fit: R-squared= .886769, Adjusted R-squared = .86318 |
| Model test: F[ 10, 48] = 37.59, Prob value = .00000 |
| Diagnostic: Log-L = -188.4047, Restricted(b=0) Log-L = -252.6651 |
| LogAmemiyaPrCrt.= 3.926, Akaike Info. Crt.= 6.759 |
| Autocorrel: Durbin-Watson Statistic = 1.56638, Rho = .21681 |
+-----------------------------------------------------------------------+
+---------+--------------+----------------+--------+---------+----------+
|Variable | Coefficient | Standard Error |t-ratio |P[|T|>t] | Mean of X|
+---------+--------------+----------------+--------+---------+----------+
Constant -23.16144958 13.072627 -1.772 .0828
COMPULSO 14.07568786 3.3015924 4.263 .0001 .30508475
ARRIVAL 3.139754254 3.3495967 .937 .3533 .10169492
CONCURR 13.08922870 2.2800452 5.741 .0000 .28813559
SEATSHAR -.6266177991E-01 .29859174 -.210 .8347 7.5910619
RESTDAY 6.134729887 3.1608517 1.941 .0582 .74576271
LIST 7.845312369 5.2345194 1.499 .1405 .38983051
CONSTITU 10.95397344 3.7786364 2.899 .0056 .25423729
NETPAYER -7.184236886 2.4629240 -2.917 .0054 .20338983
NETRECEI -2.929871650 4.0994987 -.715 .4783 .30508475
NPELEC .8664225166 .16016094 5.410 .0000 80.653898
--> REGRESS;Lhs=EPELEC;Rhs=ONE, compulso, arrival, concurr, seatshar, restday...
,constitu, netpayer, netrecei, npelec
, y1999,italcomp, luxembou, sweden
$
+-----------------------------------------------------------------------+
| Ordinary least squares regression Weighting variable = none |
| Dep. var. = EPELEC Mean= 62.69152542 , S.D.= 17.67353650 |
| Model size: Observations = 59, Parameters = 15, Deg.Fr.= 44 |
| Residuals: Sum of squares= 1028.446251 , Std.Dev.= 4.83464 |
| Fit: R-squared= .943232, Adjusted R-squared = .92517 |
| Model test: F[ 14, 44] = 52.22, Prob value = .00000 |
| Diagnostic: Log-L = -168.0363, Restricted(b=0) Log-L = -252.6651 |
| LogAmemiyaPrCrt.= 3.378, Akaike Info. Crt.= 6.205 |
| Autocorrel: Durbin-Watson Statistic = 2.13632, Rho = -.06816 |
+-----------------------------------------------------------------------+
+---------+--------------+----------------+--------+---------+----------+
|Variable | Coefficient | Standard Error |t-ratio |P[|T|>t] | Mean of X|
+---------+--------------+----------------+--------+---------+----------+
Constant -20.74887399 10.242952 -2.026 .0489
COMPULSO 14.82021308 3.3276061 4.454 .0001 .30508475
ARRIVAL 7.016536551 3.0620550 2.291 .0268 .10169492
CONCURR 8.278748372 2.0447487 4.049 .0002 .28813559
SEATSHAR .5989759096 .35900355 1.668 .1023 7.5910619
RESTDAY 7.497292883 2.8676639 2.614 .0122 .74576271
LIST -.2643223813 5.8362987 -.045 .9641 .38983051
CONSTITU 10.68081444 3.6922134 2.893 .0059 .25423729
NETPAYER -6.972413586 2.1313809 -3.271 .0021 .20338983
NETRECEI 2.744334773 3.6644548 .749 .4579 .30508475
NPELEC .8018445461 .12664129 6.332 .0000 80.653898
Y1999 -3.634952348 1.7628816 -2.062 .0452 .23728814
ITALCOMP -15.02830579 5.4641666 -2.750 .0086 .50847458E-01
LUXEMBOU 11.56416772 4.5178705 2.560 .0140 .84745763E-01
SWEDEN -11.05695089 4.6555635 -2.375 .0220 .33898305E-01