What is a “normal threshold” of invalid votes? —
English
 
Subscribe to ACE Newsletter
These topic pages provide a quick overview and easy access to all content that can be found on ACE for any give topic of interest - weather encyclopaedia files, electoral materials, comparative data, consolidated replies, case studies, or other.

ELECTORAL FRAMEWORK

Legal Framework Boundary Delimitation Electoral Systems

ELECTORAL PARTICIPATION

Gender and Elections Parties and Candidates Disability and Elections Media and Elections Civic and Voter Education Direct Democracy

ELECTORAL MANAGEMENT

Electoral Management Cost of Elections

ELECTORAL INTEGRITY

Electoral Integrity Campaign Finance Measuring Electoral Quality Election Observation Electoral Assistance

ELECTORAL OPERATIONS

Voter Registration Voting Operations Voter Identification Results Management Out of Country Voting Elections and Technology Elections and Security
Countries
Encyclopaedia Electoral Materials Comparative Data Election Calendar Observation Portal BRIDGE Extended Learning Latest Updates
Ask the Experts Consolidated Replies Find An Expert Access ACE Practitioners' Workspace
About ACE The History of ACE Contact ACE ACE Partners Practitioners' Network FAQ

What is a “normal threshold” of invalid votes?

 

The percentage of invalid votes varies between types of elections (such as presidential or parliamentary), the technology or logistics involved in the voting system (for example electronic voting or paper ballots), as well as the consistency and adherence to instructions for what is to be counted as invalid votes among polling station staff.

For instance, in the national elections held in the world since the year 2000, where data was available, an estimated average of 4.3% of votes were considered invalid (see International IDEA’s Voter Turnout Database).

If the percentage of invalid votes in an election is seen as high, it can increase the pressure for a recount of the results or even nullification and reorganization of the election.

The views of ACE Practitioner members on this topic can be found at the following consolidated reply: "Acceptable" number/proportion of invalid votes?

 

What do you think is a normal or acceptable proportion of invalid ballots in an election?

  • Between 2-4 % of votes cast.
  • Between 4-8 % of votes cast.
  • More than 8% of votes cast.
  • Don’t know.
Total votes
199 people have voted on this poll.
Document Actions

About normality

Posted by Horacio Boneo at Dec 13, 2011 11:34 AM
I have never been a fan of this section which occupies prime space in the home page and has limited response (last question received an average of less than one answer per day). The main problem is that the questions posed need a context and usually additional information to be properly answered. Take the case of this question. First, there needs to be a distinction (which may or not exist depending on the specific regulations of a country) between blank and null votes. Null votes are frequently an expression of protest - Cacareco, a rhinoceros at Sao Paulo Zoo received enough write-in votes to be elected councilperson in the municipal elections of 1958. Second, diachronic comparisons within a given society tend to be more relevant than international comparisons. There is a “normal” percentage of invalid votes in a given society that might be used to say that the percentage is a given election is “abnormally high” or viceversa. Changes in voting methods or in voter education can be evaluated in terms of their impact on the number of invalid votes in relation to a normality defined in previous elections. On the other hand, international comparisons mix systems which use voting machines which may not have “invalid” votes and manual systems with a complicated system of preferences. It will mix societies with a high level of illiteracy with others with a predominance of well-educated citizens.

Invalid votes “normal threshold”

Posted by Maria Helena Alves at Apr 27, 2012 10:29 AM
I apologize for joining this discussion at a rather late time, but having just read the question and the answers I felt that I needed to support Horacio's comment on this particular issue. I also do believe that the question of "invalid" votes has to be analysed / explained in the context of a specific election, country and respective election system.

Invalid Votes recommanded

Posted by Juma Thomas Otieno at Jan 09, 2012 12:44 PM
In as much as your objectives narrow us down to choose, I still believe that if every process towards a free and fair elections are done, the invalid votes should be at the very minimum percentage. This caters for known and unintentional errors.

ProbleM

Posted by Daniel at Jan 18, 2012 05:47 PM
The real problem comes when the number of invalids is bigger than the difference between the candidates. This is possible mainly with FPTP.