Why is Turnout Higher in Some Countries than in Others?

Appendices

Appendix B – Institutional Variables

North America

Country	Compulsory voting with penalties	Compulsory voting without penalties	PR and mixed compensatory	Compulsory registration	Polling day registration	Initiative/ State	Leave of absence	Advance poll	Mail-in voting	Proxy voting	Easy access to voting
Bahamas								x			0.33
Barbados											0
Belize						x				x	0.33
Canada					x	x		x	x		0.66
Costa Rica		x	x	x		x	x				0
Jamaica						x		x			0.33
Panama		x		x		x	x				0
St. Lucia			x					x			0.33
St. Vincent  and the  Grenadines											0
Trinidad  and Tobago						x		x			0.33

South America

Country	Compulsory voting with penalties	Compulsory voting without penalties	PR and mixed compensatory	Compulsory registration	Polling day registration	Initiative/ State	Leave of absence	Advance poll	Mail-in voting	Proxy voting	Easy access to voting
Argentina	x		x			x	x				0
Bolivia	x		x	x			x				0
Brazil	x		x	x			x				0
Chile	x						x				0
Ecuador	x					x	x				0
Guyana			x	x		x		x		x	0.66
Uruguay	x		x	x	x		x				0
Venezuela		x	x	x		x	x				0

Asia

Country	Compulsory voting with penalties	Compulsory voting without penalties	PR and mixed compensatory	Compulsory registration	Polling day registration	Initiative/ State	Leave of absence	Advance poll	Mail-in voting	Proxy voting	Easy access to voting
Bangladesh						x			x		0.33
Cyprus	x		x	x		x	x				0
India						x	x		x		0.33
Israel			x		N/A	x	x	x			0.33
Japan					N/A	x	x	x	x		0.66
Mongolia						N/A	x				0
Philippines	x			x			x				0
Taiwan				N/A	N/A	N/A	x	N/A	N/A	N/A	N/A

Europe

Country	Compulsory voting with penalties	Compulsory voting without penalties	PR and mixed compensatory	Compulsory registration	Polling day registration	Initiative/ State	Leave of absence	Advance poll	Mail-in voting	Proxy voting	Easy access to voting
Belgium	x		x			x	x			x	0.33
Bulgaria			x			x	x				0
Czech  Republic			x		N/A	N/A					0
Denmark			x			x		x			0.33
Estonia			x		N/A	N/A	x	x	x		0.66
France				x			x			x	0.33
Germany			x			x	x		x		0.33
Hungary			x		x	x	x				0
Ireland			x			x		x	x		0.66
Italy		x	x		N/A	N/A	x				0
Latvia			x		N/A	N/A	x		x		0.33
Lithuania					x	x	x	x	x		0.66
Luxembourg	x		x		N/A	x	x		x		0.33
Malta			x			x	x	x			0.33
Netherlands			x			x			x	x	0.66
Poland			x		x	x	x				0
Portugal		x	x	x			x	x			0.33
Romania			x	x		x	x				0
Slovakia			x		x	x					0
Slovenia			x	N/A	N/A	N/A	x	x	x		0.66
Spain			x	x		x	x		x		0.33
Sweden			x			x	x	x	x	x	1
United  Kingdom						x			x	x	0.66

Oceania

Country	Compulsory voting with penalties	Compulsory voting without penalties	PR and mixed compensatory	Compulsory registration	Polling day registration	Initiative/ State	Leave of absence	Advance poll	Mail-in voting	Proxy voting	Easy access to voting
Australia	x			x			x	x	x		0.66
New  Zealand			x	x	x		x	x	x		0.66
Papua  New Guinea				x					x		0.33
Vanuatu						N/A	x			x	0.33

  N/A: not available

Appendix C – Countries, Elections and Turnout

Africa

Country/Election	Turnout based on the number of people registered on the lists of electors	Turnout based on the population old enough to vote
Benin 1995	75.8	73.7
Benin 1991	51.7	46.5
Cape Verde 2001	54.1	N/A
Cape Verde 1995	76.5	79.5
Cape Verde 1991	75.3	71.8
Madagascar 1998	60.1	40.6
Madagascar 1993	60.0	51.4
Malawi 1994	80.0	67.7
Mali 1992	21.1	21.9
Namibia 1999	62.8	N/A
Namibia 1994	76.0	63.8
Sao Tome 1998	64.7	N/A
Sao Tome 1994	52.1	49.5
Sao Tome 1991	76.7	69.6
South Africa 1999	89.3	N/A
South Africa 1994	N/A	85.5

North America

Country/Election	Turnout based on the number of people registered on the lists of electors	Turnout based on the population old enough to vote
Bahamas 1997	90.0	66.7
Bahamas 1992	91.9	68.5
Barbados 1999	63.1	N/A
Barbados 1994	60.3	66.0
Barbados 1991	63.7	67.4
Belize 1998	90.1	N/A
Belize 1993	74.6	68.8
Canada 2000	61.2*	N/A
Canada 1997	67.8	56.2
Canada 1993	69.6	63.9
Costa Rica 1998	70.3	73.7
Costa Rica 1994	81.2	84.2
Costa Rica 1990	82.5	85.1
Jamaica 1997	65.4	48.8
Jamaica 1993	66.7	44.1
Panama 1999	75.9	N/A
Panama 1994	73.7	70.1
St. Lucia 1997	66.0	N/A
St. Lucia 1992	62.8	76.8
St. Vincent and the Grenadines 2001	68.7	N/A
St. Vincent and the Grenadines 1998	67.4	75.9
St. Vincent and the Grenadines 1994	65.6	73.2
Trinidad and Tobago 2000	63.1	N/A
Trinidad and Tobago 1995	63.3	67.3
Trinidad and Tobago 1991	65.8	70.4

South America

Country/Election	Turnout based on the number of people registered on the lists of electors	Turnout based on the population old enough to vote
Argentina 1999	78.6	N/A
Argentina 1995	81.0	79.8
Argentina 1993	79.7	78.1
Argentina 1991	89.7	89.4
Bolivia 1997	70.0	62.3
Bolivia 1993	72.2	50.0
Brazil 1994	82.2	76.8
Brazil 1990	84.6	76.6
Chile 1997	87.3	N/A
Chile 1993	91.0	81.9
Ecuador 1998	47.3	48.5
Ecuador 1996	67.9	67.8
Ecuador 1994	65.5	66.3
Ecuador 1990	69.4	64.7
Guyana 2001	89.1	N/A
Guyana 1997	88.4	80.2
Uruguay 1999	91.7	N/A
Uruguay 1994	91.4	96.1
Venezuela 1998	52.6	52.6

Asia

Country/Election	Turnout based on the number of people registered on the lists of electors	Turnout based on the population old enough to vote
Bangladesh 1996	75.6	64.6
Bangladesh 1991	55.4	61.5
Cyprus 2001	90.5	N/A
Cyprus 1996	90.1	75.9
Cyprus 1991	94.3	78.6
India 1999	59.7	N/A
India 1998	62.0	N/A
India 1996	57.9	61.1
Israel 1999	78.7	N/A
Israel 1996	79.3	84.7
Israel 1992	77.4	81.7
Japan 2000	60.6	N/A
Japan 1996	59.0	59.8
Japan 1995	44.5	44.9
Mongolia 2000	82.4	N/A
Mongolia 1996	88.4	73.6
Philippines 1998	78.7	N/A
Philippines 1995	70.7	68.4
Taiwan 1998	68.1	N/A
Taiwan 1996	76.2	75.1

Oceania

Country/Election	Turnout based on the number of people registered on the lists of electors	Turnout based on the population old enough to vote
Australia 1998	95.2	N/A
Australia 1996	95.9	82.5
Australia 1993	95.6	83.4
Australia 1990	95.5	82.1
New Zealand 1999	83.1	N/A
New Zealand 1996	88.3	83.0
Papua New Guinea 1997	65.7	98.8
Papua New Guinea 1992	81.2	79.2
Vanuatu 1998	61.8	N/A
Vanuatu 1995	72.4	78.6
Vanuatu 1991	71.3	70.7

Europe

Country/Election	Turnout based on the number of people registered on the lists of electors	Turnout based on the population old enough to vote
Belgium 1999	90.6	N/A
Belgium 1995	91.1	83.2
Belgium 1991	92.7	85.1
Bulgaria 1997	58.9	66.9
Bulgaria 1994	75.2	81.0
Bulgaria 1991	83.9	84.5
Czech Republic 1998	74.0	76.7
Czech Republic 1996	76.3	77.6
Denmark 1998	85.9	83.1
Denmark 1994	84.3	81.7
Denmark 1990	82.8	80.4
Estonia 1999	57.4	N/A
Estonia 1995	68.9	48.8
France 1997	68.0	59.9
France 1993	68.9	61.29
Germany 1998	83.0	N/A
Germany 1994	79.0	72.4
Germany 1990	77.8	73.1
Hungary 1998	56.7	59.9
Hungary 1994	68.9	69.4
Hungary 1990	75.5	75.9
Ireland 1997	66.1	66.7
Ireland 1992	68.5	73.7
Italy 2001	81.2	N/A
Italy 1996	82.9	87.4
Italy 1994	86.1	90.8
Latvia 1998	71.9	N/A
Latvia 1995	71.9	50.6
Lithuania 2000	58.6	N/A
Lithuania 1996	52.9	50.0
Lithuania 1992	75.2	70.2
Luxembourg 1999	86.5	N/A
Luxembourg 1994	88.3	60.5
Malta 1998	95.4	N/A
Malta 1996	97.2	98.0
Malta 1992	96.0	95.3
Netherlands 1998	73.2	N/A
Netherlands 1994	78.7	75.2
Poland 1997	47.9	48.8
Poland 1993	52.1	52.0
Poland 1991	43.2	44.4
Portugal 1999	61.0	N/A
Portugal 1995	66.3	79.1
Portugal 1991	68.2	77.7
Romania 2000	65.3	N/A
Romania 1996	76.0	78.2
Slovakia 1998	84.2	N/A
Slovakia 1994	75.4	75.9
Slovenia 2000	70.4	N/A
Slovenia 1996	73.3	75.5
Slovenia 1992	85.9	85.5
Spain 2000	68.7	N/A
Spain 1996	78.1	80.6
Spain 1993	77.0	77.4
Sweden 1998	81.4	N/A
Sweden 1994	88.1	83.6
Sweden 1991	86.7	82.8
United Kingdom 2001	59.4	N/A
United Kingdom 1997	71.5	69.4
United Kingdom 1992	77.8	75.4

N/A: not available

 *The turnout of 61.2% in 2000 was adjusted to arrive at the final turnout of 64.1%, after our normal maintenance of the National Register of Electors to remove the names of deceased electors and duplicates arising from moves. The Chief Electoral Officer of Canada explained the adjustment during his appearance before the Subcommittee on Electoral Boundaries Readjustment on October 6, 2003, and his appearance to discuss the 2004 Main Estimates before the Standing Committee on Procedure and House Affairs on March 5, 2004.

Appendix D – Understanding Regression Analyses

Tables 1A and 3B show the results of multi-varied regression analyses of factors that affect the voter turnout measured as a function of the number of registered electors or of the voting age population. Multi-varied regression analysis is used to identify the specific and independent effect of a given factor once the other factors included in the analysis have been taken into account. Regression analysis is used to determine, for example, whether, all other things being equal, that is, an equivalent level of socio-economic development, population and geographic location, countries with a high population density have a higher turnout than countries with a low population density.

In the tables, an asterisk is used to indicate whether a given variable has a statistically significant effect on turnout. The absence of an asterisk indicates that it cannot be concluded that the factor independently affects voter turnout. For example, the results presented in tables 1A and 1B indicate that population density does not appear to have a specific impact on turnout. The presence of an asterisk signifies that there is a 90 percent chance that the variable has an effect, in other words, that there is only one chance in 10 that the observed relation is random. Two asterisks means that the probability that there is "really" an effect is 95 percent, and three asterisks means that the probability is 99 percent.

Thus, one or more asterisks allow us to conclude that a factor affects turnout. The regression coefficient indicates the extent of that influence. It is useful in this regard to distinguish among certain types of variables. First, the geographic variables. For the purposes of the analysis, we used Western Europe as a reference point.

The data in Table 1A indicate that, all other things being equal, that is, once the country's level of development and size have been accounted for, the participation rate is typically 8 percentage points lower in North America and Eastern Europe than in Western Europe. Then there are the dichotomous variables, used in particular to measure institutional factors. The "compulsory voting with penalties" variable, for example, has a value of 1 in countries with compulsory voting with penalties and a value of 0 in countries without it.

Table 2A indicates that, all other things being equal, turnout is 12.6nbsp;percentage points higher in countries with compulsory voting with penalties than in countries without it. The reader should note that the institutional variable "ease of voting" is handled differently. On the numerical scale, a country has a maximum score of 1 if it is possible to vote in advance, by mail or by proxy, of 0.66 if two of the three options are available, of 0.33 if one is, and 0 if none are.

The results in Table 3A indicate that, all other things being equal, the turnout in a country with a score of 1 is 11 percentage points higher than that of a country with a score of 0. There are, finally, numerical variables such as the size of the population and the per capita GDP. The coefficients associated with these variables cannot be easily interpreted, since the relation is logarithmic, which means that most of the difference is at the "bottom" of the scale and that, for example, the turnout is particularly low in the poorest countries while the gap between moderately rich and rich countries is minimal.

• Previous
• Table of Contents