View all publications

1 Introduction

1.1 Objectives

The broad objectives of the research were to determine:

  • whether households are willing to pay for the safe disposal of electronic and electrical items
  • how much households are willing to pay
  • the preferred timing of any fee (at time of disposal or when purchasing a new item)
  • preferences for collection options.

1.2 Method

The results in this report are based on questions asked in a UMR Research nationwide omnibus survey. This was a telephone survey of a nationally representative sample of 750 New Zealanders, 18 years of age and over. Field work was conducted from 6 to 9 April 2006 at UMR Research's national interview facility in Auckland.

The sample

Telephone numbers for those included in the sample were sourced from the Telecom White Pages. The sample universe was stratified into 23 telephone directory regions. The number of residential addresses in each of these regions was determined, and a quota was specified for the proportion of the sample that must fall in each region. Table 1 shows the percentage of respondents for each area and the number of respondents to be surveyed in each area out of the sample of 750 respondents.

Table 1: Stratified sample, by telephone region

Directory region Geographic region percentage Sample size

01

Auckland 1

5.1 per cent

38

02

Auckland 2

4.1 per cent

31

03

Auckland 3

7.4 per cent

55

04

Auckland 4

7.4 per cent

55

05

Auckland 5

4.5 per cent

34

06

Auckland 6

2.1 per cent

16

07

Bay of Plenty

7.1 per cent

53

08

Christchurch

12.3 per cent

92

09

Gisborne

1.0 per cent

7

10

Hawke's Bay

4.0 per cent

30

11

Manawatu

3.6 per cent

27

12

Marlborough

1.2 per cent

9

13

Nelson & Bays

2.5 per cent

19

14

Northland

3.4 per cent

25

15

Otago

4.7 per cent

35

16

Southland

2.7 per cent

20

17

Taranaki

2.8 per cent

21

18

Timaru & Oamaru

2.2 per cent

17

19

Waikato

8.1 per cent

61

20

Wairarapa

1.3 per cent

10

21

Wanganui

1.6 per cent

12

22

Wellington

10.3 per cent

77

23

West Coast

0.9 per cent

6

Total

100 per cent

750

The telephone sampling scheme

The random sample of telephone numbers was generated from all number ranges found in the Telecom White Pages for New Zealand. Random digit dialing was conducted from this sample so that unlisted numbers were captured in the sample.

To limit the sample frame to private households with telephones, the following types of telephone numbers were filtered out from the sample:

  • Telecom Yellow Pages
  • disconnected or fax lines
  • where the interviewer determined the contact was not a private household
  • business lines.

Margin of error

The margin of error for a 50 per cent figure at the 95 per cent confidence level for the sample of 750 is plus or minus 3.58 per cent.

Call-back and calling times

Up to five call-backs were made to initially selected respondents to ensure that non-response had a minimal impact on the representativeness of the sample. Appointments were made to ring back respondents if the time they were first contacted was not convenient.

Respondents were called between 5.30 pm and 9.00 pm during the week, between 9.30 am and 6.00 pm on Saturday and between 9.30 am and 9.00 pm on Sunday.

Response rate and weighting regime

The response rate for this sample was 36 per cent.

The survey data were rim weighted by age, sex and household size. Rim weighting calculates weights using a form of regression analysis. This requires two parameters: a limit, which defines how close the weighting procedure must get to the targets that have been set in order for the weights to be acceptable, and the number of iterations, which defines the number of times the weight calculations may be repeated in order to reach the cell targets.

Rim weighting is designed to weight characteristics simultaneously. The rim weighting process achieves all the desired proportions while distorting each variable as little as possible.