• About this site|
• Site index|
• Contact us|
• Jobs

Household Sustainability Survey 2008

Executive summary

1 Introduction and objectives

2 Survey methodology

3 Overview of results

4 Population segmentation

2 Survey methodology

In this section of the report, we outline the methodology that was used to complete the survey monitor.

2.1 Overview

A total sample of 1000 New Zealanders, aged 15 years and over, was interviewed for the monitor survey. The sample was randomly selected using the General and Māori Electoral Rolls, with telephone numbers obtained from Telecom New Zealand’s White Pages.

Interviewing was conducted by telephone, between 3 and 27 September 2008, from Research New Zealand’s purpose-built CATI (Computer Assisted Telephone Interviewing) enabled call centre. This call centre is Interviewing Quality Standards (IQS)-accredited. Before being interviewed for the survey, all respondents were sent a pre-notification letter introducing them to the research.

The overall response rate for the survey, calculated based on a method approved by Statistics New Zealand, was 27.6 percent.

Given that respondents knew the Ministry for the Environment was the sponsor of the research, it is likely that some response bias is present⁴.

2.1.1 Sampling design

The survey sample was selected to be representative of the New Zealand population 15 years of age or more. The sample was sourced from the General and Māori Electoral Rolls, using a multi-random procedure:

• At the first step, eligible electors were selected at random from the rolls in order to identify households that could be approached.

• When interviewers contacted these households, they established the total number of qualifying respondents usually resident in the household (i.e. 15 years of age or more) and randomly selected from these people (using the ‘next birthday’ technique).

2.1.2 Survey design

The design of the questionnaire for the survey monitor took into account three elements:

• Last year’s original benchmark survey.

• The Defra question framework, comprising 17 question items.

• The results of an online survey completed to identify sustainable behaviours that members of the general public believed would be easy and they would also be willing to adopt.

The draft survey questionnaire was piloted before being finalised.

2.2 Interviewing method

2.2.1 Pre-notification

All respondents/households that were randomly selected from the General and Māori Electoral Rolls were sent a pre-notification letter on the Ministry for the Environment’s letterhead introducing the survey. In accordance with the Code of Practice of the Market Research Society of New Zealand Inc., this letter outlined the purpose of the survey, how they had been selected and what participation in the survey would involve.

The letter also stated that they/their household might shortly be contacted by an interviewer from Research New Zealand, that participation in the survey was voluntary and that, if an interview was conducted, it would be completed on a confidential basis.

Frequently Asked Questions were included on the reverse of the letter and our 0800 number was provided in case people had additional questions, or they wanted to ‘opt in’ or ‘opt out’ of the survey.

2.2.2 Telephone interviewing

All interviewing for the monitor survey was completed from our purpose-built, CATI-enabled call centre. This means that the survey questionnaire appeared on interviewers’ computer screens and they immediately entered respondents’ answers to questions into our computer system. This ensures human error is minimised and the survey results are more immediately available.

2.3 Analysis and reporting

2.3.1 Weighting

The survey data has been weighted by age and gender to ensure that the results quoted at the total sample level are truly representative of the New Zealand population aged 15 years and over. The weighting parameters were sourced from Statistics New Zealand and based on the 2006 Census of Population and Dwellings.

2.3.2 Analysis and segmentation

As noted in the previous section, respondents have been segmented using the approach taken for the United Kingdom’s Department for Environment, Food and Rural Affairs’ (Defra) Survey of Public Attitudes and Behaviours Toward the Environment 2007.

In Defra’s study, the British general public were segmented using 44 different variables. It was not possible, nor desirable, to include all 44 variables in the New Zealand survey. A subset of 17 key questions had been identified by the research company BMRB Social Research (part of BMRB Limited: British Market Research Bureau) as being the most appropriate questions to include in Defra’s future surveys to recreate the segmentation. This subset was used for the New Zealand survey.

All of the survey respondents were asked if they were happy with their current lifestyle in relation to the environment, or if they wanted to be doing a bit or a lot more to help care for the environment. In addition, they were asked to rate their levels of agreement with each the following 16 statements (this set thus comprises 17 key questions):

• The Earth has very limited room and resources.

• If things continue on their current course, we will soon experience a major environmental disaster.

• We are close to the limit of the number of people the Earth can support.

• The so-called environmental crisis facing humanity has been greatly exaggerated.

• It’s not worth New Zealand trying to combat climate change because other countries will just cancel out what we do.

• It’s only worth doing environmentally friendly things if they save you money.

• The effects of climate change are too far in the future to really worry me.

• It’s not worth me doing things to help the environment if others don’t do the same.

• It would embarrass me if my friends thought my lifestyle was purposefully environmentally friendly.

• I find it hard to change my habits to be more environmentally friendly.

• I don’t pay much attention to the amount of water I use at home.

• People have a duty to recycle.

• I don’t really give much thought to saving energy in my home.

• I would only travel by bus if I had no other choice.

• People who fly should bear the cost of the environmental damage that air travel causes.

• For the sake of the environment, car users should pay higher taxes.

In order to recreate the Defra segmentation for analysis purposes, an algorithm that was developed by BMRB Social Research was then used. This involved applying various weights or multipliers (known as classification function coefficients) to each respondent’s answers to the key statements noted above: the calculated index scores indicated the probability of a respondent belonging to a specific segment.

Subsequent validation of the accuracy of the segmentation was completed using Discriminant Analysis of the respondents’ un-weighted answers to the 17 key questions. This analysis identified that the algorithm had accurately classified 83 percent of the respondents in their respective segments⁵.

Examination of the data identified that the misclassifications of respondents resulted primarily from two main sources. One was that the algorithm places the greatest weight on statements which most of the respondents agreed to people have a duty to recycle (92 percent of all of the New Zealand respondents agreed or strongly agreed with the statement).

The second main source of misclassification was that for many respondents the algorithm calculated quite similar probabilities of their being in more than one segment. Such respondents’ un-weighted answers were examined along with their probability scores calculated during the Discriminant Analysis (to examine respondents’ various key attitudes towards the environment and whether these were most similar to the members of their original segment or those of another segment). Where it was clear that these respondents belonged in an alternative segment, they were reallocated as appropriate⁶.

After correcting for misallocations, the distribution of respondents in relation to their final segments was as follows:

• Positive Greens – 14 percent.

• Waste Watchers – 39 percent.

• Concerned Consumers – 18 percent.

• Sideline Supporters – 5 percent.

• Cautious Participants – 8 percent.

• Stalled Starters – 5 percent.

• Honestly Disengaged – 11 percent.

2.3.3 Margins of error

As a result of the weighting used in this survey, the actual margin of error is slightly more than the conventional calculation (i.e. with a simple random sample).

The maximum margins of error for the total sample and each of the segments (at the 95 percent confidence level) are listed in Table 1.

Table 1: Margin of Error (Weighted)

For example, 52 percent of the total sample reported that I do quite a few things that are environmentally friendly; based on the maximum margin of error of ±3.4 percent for the total sample we are 95 percent confident that between 48.6 percent and 55.4 percent of New Zealanders feel this way.

Similarly, the survey found that 50 percent of the sub-sample that we have identified as being Positive Greens reported the same thing, based on the margin of error for the sub-sample of 137 respondents (±9.2 percent) we are 95 percent confident that between 40.8 percent and 59.2 percent of New Zealanders who are ‘Positive Greens’ feel this way.

Note that the margin of error for estimating the sizes of the segments among the general public is no more than the maximum margin of error for the total sample (a maximum of ±3.4 percent).

2.3.4 Structure of report

This report is structured in five sections as follows:

(Executive summary)

1. Introduction, background to and objectives of the current monitor
2. Methodological approach
3. Overview of key findings for the total sample
4. Demographic, lifestyle, general attitudes, and behavioural descriptions of the different segments.

Please note that significant differences in tabulations throughout this report are shown as follows:

• Results in blue (bold) indicate a significantly higher figure compared to the total sample.

• Results in red (italics) indicate a significantly lower figure compared to the total sample.

2.3.5 Demographic profile of respondents

Table 2 provides a weighted and un-weighted demographic profile of the survey respondents.

Table 2: Respondent demographic profile

Totals may not sum to 100% due to rounding.

[ Previous| Next