@ All data are originally from Kaare Strom. See his 1985 APSR article for further information. We have corrected a number of errors and added variables. These data were used for King/Alt/Burns/Laver's 1990 AJPS article. To replicate the results in that article, note that the dependent variable is durat, and the explanatory variables in the final model are invest fract polar numst2 format eltime2 caretk2. Be especially careful about the method of censoring. We created a variable (from those below) which is 1 if the cabinet is 12 months or less from CIEP (censored), and 0 otherwise. The following are the variables and their codes for the CIEP dataset. There are no records with missing data. Belgium 1-yes 0-no Canada 1-yes 0-no Denmark 1-yes 0-no Finland 1-yes 0-no France 1-yes 0-no Iceland 1-yes 0-no Ireland 1-yes 0-no Israel 1-yes 0-no Italy 1-yes 0-no Nether (Netherlands) 1-yes 0-no Norway 1-yes 0-no Portug (Portugal) 1-yes 0-no Spain 1-yes 0-no Sweden 1-yes 0-no UK 1-yes 0-no durat duration in months coded in number of months crisis duration of preceding crisis in days coded in number of days singpar cabinet composition 1-single party, 0-coalition nonpart cabinet composition 1-non-partisan or other, 0-single party or coalition parlbas parliamentary basis coded in percentages extsup external support 1-external support, 0-no external support format number of formation attempts 1-7-number of formation attempts, 8-eight or more formation attempts eltime2 timing of government in electoral cycle 1-after election, 0-else eltimeb timing of government in electoral cycle 1-before election, 0-else eltimen timing of government in electoral cycle 1-not adjacent to election, 0-else electc cause of resignation 1-electoral, 0-government, coalitional, parliamentary- confidence, parliamentary-policy, systemic, social crisis, personal, constitutional electm mode of resignation 1-technical, for elections 0-defeat or crisis or voluntary numst2 numerical status 1-majority, 2-else caretk2 caretaker role 1-caretaker, 0-not caretaker popinfl influence of opposition in parliament 1-low, 2-high invest investiture requirement 1-has investiture requirement, 0-no investiture requirement oppconc strength of major section of opposition in percentage divided by 1-parliamentary basis in percentage oppcon2 strength of major section of opposition in percentage divided by 100-parliamentary basis in percentage oppcond 1-if oppcon2=1, 0-else fract Rae's fractionalization index (seats) np number of effective parties 1/[1-fract*10^(-3)] oppinfl influence of opposition 1-low, 2-medium, 3-high ropinfl revised oppinfl polar polarization index prox proximity ident identifiability 1-low, 2-medium, 3-high volat volatility (seats) response electoral responsiveness when government is formed 1-losers only, 2-split, majority loser, 3-even split, 4-split, majority winner, 5-winners only decprox decade proximity salien electoral salience (prox/.85) + (decprox/.24) t length of CIEP coded in years ciep12 constitutional interelection period influence if govt. falls within 12 months of the CIEP 0-if durat > (t*12 - 12), 1-else ciep24 constitutional interelection period influence if govt. falls within 24 months if the CIEP 0-if durat > (t*12 - 24), 1-else cieptw 1=12 months or fewer, 0 otherwise (used by KABL). @ input coal.asc; invar durat belgium canada denmark finland france iceland ireland israel italy nether norway portug spain sweden uk t popinfl oppinfl ropinfl prox ident volat response invest polar fract numst2 parlbas crisis format oppconc eltime2 eltimeb eltimen caretk2 nonpart electc electm singpar extsup salien oppcon2 oppcond np decprox ciep12 ciep24; output coal; outtyp f;