Methodology

For more information contact: Dr Helena Catt, Chief Executive, Electoral Commission, catt@elections.govt.nz, tel +64 +4 +474 0676.

Summary

The levels of understanding fit within historic patterns.  The question wording has been changed overtime so that some differences will be due to these changes.  There is a recurrent cycle of the number saying that they don’t know decreasing after the election.  The overall trend is that over time the numbers giving the wrong answer has declined. 

The results fit the normal cyclical pattern and do not raise any immediate concerns.

Which is the most important vote?

Just over half correctly identified the party vote as the one that was most important in determining the number of MPs each party will have in parliament.  The second most popular answer was ‘both party and electorate vote’, probably reflecting the importance that winning an electorate had in this election for two of the parliamentary parties and the overhang resulting from the Maori Party’s party vote (Table 1).

Overall the results indicate good levels of understanding of the importance of the party vote.  Knowledge fluctuates in relation to the proximity of an election and indicates a reaction to the election specific context (Table 2). 

The Threshold

Post election, over half of the respondents recall that a party must win  5% of the party vote, an increase on the pre-election level.  Nearly a quarter say that they do not know the answer (Table 3).

The level of those saying that they don’t know has fallen markedly in the past decade and is at the lowest point to date (Table 4).

The Difference Between Voters and Non-Voters

To both of the questions on understanding of MMP, non-voters are much more likely than voters to say that they do not know the answer.  Voters are more likely to give the correct answers to both questions (Tables 6 & 7).

Other research

Data from the New Zealand Election Study which repeats these questions for a much larger sample will allow more detailed analysis.  Reports will be produced during 2006 as data and analysis is available.

Which is the most important vote?

 ‘From what you know and have heard, which of the two votes that you have in MMP is more important in deciding the number of MPs each party will have in parliament?  

• The party vote,
• The electorate vote’

Table 1  Most important vote, pre and post election

The correct answer is party vote, which was the answer of the majority of respondents after the election. However, given the role played by electorate votes in the 2005 election, with ACT and United Future New Zealand crossing threshold because of the electorate vote and Maori Party gaining more electorate MPs than their party vote allocation gave, then ‘both are important’ is a valid answer.  The large increase in the proportion answering ‘both’ between the pre and post election surveys indicates a reaction to the specifics of the election campaign. 

Less than a tenth did not answer the question, indicating widespread familiarity with MMP and a willingness to answer questions about it. 

Previous surveys used a slightly different question, the 2003 version is given above table 2 although there have been some ‘tweaks’ over the preceding years.  Some of the changes may be due to the change in question which means people no longer have to remember the name of the vote. 

Table 2 – most important vote 1995-2005

Just judging from what you know and have heard, which of those votes is more important in deciding the number of MPs each party will have in Parliament?’

a = pre-election   b= post-election

Those who did not answer or said ‘don’t know’ were probed for an answer and these are included in the results shown.  Respondents could only give one answer so there is no equivalent measure for those who said ‘both’ in 2005.   Prior to 1999 the number who gave a different answer was not recorded in the survey reports so they are included with the ‘don’t know’ responses.  The break in each line on the graph indicates the major change in the question wording.

As electorate vote is always the wrong answer and the one that people might give if they were still thinking about the old FPP system, then this provides the most reliable overtime measure of understanding.  Whilst there is some fluctuation, the overall trend is downwards.  In the post election survey less than a tenth held this view.

The proportion of people saying that they ‘don’t know’ or that do not answer also indicates changing levels of understanding of MMP. These levels fluctuate markedly with a clear cycle related to the election and thus a focus of attention of the electoral system.  The lines for ‘don’t know’ and the correct answer of ‘party vote’ mirror each other indicating that those who seek the information anew for each election are gaining the correct information. 

Overall the results indicate good levels of understanding of the importance of the party vote.  Knowledge fluctuates in relation to the proximity of an election and indicates a reaction to the election specific context. 

The Threshold

 ‘Party votes for a party are used to allocate seats in parliament for all parties which cross the threshold. Can you recall which of these a party has to do in order to cross that threshold? 

• Win 5 percent of all party votes,
• Win one electorate,
• either of these’

Table 3  What is the threshold – pre and post election

The correct answer is either.  Post election, over half of the respondents recall that a party must win  5% of the party vote, an increase on the pre-election level.  Nearly a quarter say that they do not know the answer.  This level is much higher than for the earlier question indicating the clear hierarchy in which details of MMP are widely known.  Whilst all need to know the importance of the party vote in order to cast an effective vote, many voters will not feel that the threshold impacts on their decision. 

Previous surveys used a different question. The 2003 version is given above Table 6 (although there were ‘tweaks’ over the preceding years).  Some of the changes in response levels may be due to the change in question which means people no longer have to remember the name of the votes. In the earlier question if people answered ‘5% of party vote’ and ‘win and electorate MP’ then both of these answers were recorded.  The table distinguishes between those who mentioned both aspects and those who mentioned just one.  The break in each line on the graph indicates the major change in the question wording. 

Table 4 What is the threshold – 1995-2005

‘From what you have heard, what must a party do to qualify for list MPs?’

a = pre-election   b= post -election

The top line (triangles) shows the percentage who said that they did not know and the other line shows those who correctly named both parts of the threshold

Whilst the level of those saying that they don’t know is still high it has fallen markedly in the past decade and is at the lowest point to date.

The top line (squares) shows the percentage who mentioned only 5% of the party vote and the other line shows those who only mentioned winning an electorate.

Consistently people are more aware of the 5% of party vote part of the threshold.

How easy is MMP?

‘ Thinking about the system we use for voting in elections – MMP.  How easy do you think it is for people like you to understand MMP?

•             (5 point scale)’

Table 5 How easy is MMP – pre and post election

Respondents are pretty evenly split between those who think it is easy, those who think it is difficult and the rest.  This question was not asked in the past so no historical comparison is possible.

The Difference Between Voters and Non-Voters

In the post-election survey people were asked if they had voted at the election so the understanding of MMP amongst voters and non-voters can be compared.  As the number of non-voters is small the margin of error for the non-voters is twice as big as for the voters so these tables should be treated as indicative only.

To both of the questions on understanding of MMP, non-voters are much more likely than voters to say that they do not know the answer.  In both cases voters are  more likely to give the correct answer.  What we do not know from this survey is the direction of causality: are people less likely to vote because they do not think that they understand MMP or do they not know because they have already decided not to vote and therefore do not feel the need to retain information about MMP?  There is little difference between the voters and non-voters in how easy or difficult they think MMP is to understand.

The data from the New Zealand Election Study which repeats these questions for a much larger sample will allow more detailed analysis.

Table 6 Most important vote – voters and non voters

Table 7 What is the threshold – voters and non voters

Table 8 How easy is MMP – voters and non voters

Sample breakdown

Table 9 – Age of respondents

Table 10 - Gender of respondents