Statistics New Zealand gives no warranty that the information or data supplied contains no errors. However, all care and diligence has been used in processing, analysing and extracting the information. Statistics New Zealand shall not be liable for any loss or damage suffered by the customer consequent upon the use directly, or indirectly, of the information supplied in this product.
Any table or other material published in the following pages of this report may be reproduced and published without further licence, provided that it does not purport to be published under Government authority and that acknowledgement is made of this source.
Published in April 2002
Statistics New Zealand
Te Tari Tatau
Wellington, New Zealand
Catalogue Number 02432.0001
ISBN 0-478-20796-4
1. Introduction
2. Survey Objectives
2.1 Overall Objectives
2.2 Detailed Objectives
3. Population
3.1 Target Population
3.2 Survey Coverage
4. Classifying Proficiency
5. Survey Methodology
5.1 Survey Design
5.2 Sample Design
5.3 Reliability of Estimates
6. Response Rate
7. Fieldwork
7.1 The Fieldforce
7.2 Interviewer Training
Appendix One: Survey Tables
Table 1: Speaking Proficiency by Age and Sex
Table 1a: Absolute Sample Errors for Table 1 Speaking Proficiency by Age and Sex
Table 1b: Relative Sample Errors for Table 1 Speaking Proficiency by Age and Sex
Table 2: Speaking Proficiency by Age and Region
Table 2a: Relative Sample Errors for Table 2 Speaking Proficiency by Region
Table 2b: Absolute Sample Errors for Table 2 Speaking Proficiency by Region
Table 3: Speaking Proficiency by Age and Exposure to the Māori Language as a Child
Table 4a: Speaking Proficiency by Sources of Māori Language Input (up to primary school age)
Table 4b: Speaking Proficiency by Sources of Māori Language Input (from primary school age to school leaving age)
Table 5a: Speaking Proficiency by Age and Main Childhood Language (up to primary school age)
Table 5b: Speaking Proficiency by Age and Main Childhood Language
(from primary school age to school leaving age)
Table 6a: Speaking Proficiency by Age and Main Language Used by Adults (up to primary school age)
Table 6b: Speaking Proficiency by Age and Main Language Used by Adults (primary school age to school leaving age)
Table 7: Speaking Proficiency by Age and Method of Acquisition
Table 8a: Speaking Proficiency by Age and Response to Census Question
Table 8b: Response to Census Question in the MLS by Age Against the Response to the Census Language Question
Table 9: Listening Proficiency by Age and Sex
Table 10: Listening Proficiency by Age and Region
Table 11: Listening Proficiency by Age and Exposure to the Māori Language as a Child
Table 12a: Listening Proficiency by Sources of Māori Language Input (up to primary school age)
Table 12b: Listening Proficiency by Sources of Māori Language Input (from primary school age to school leaving age)
Table 13a: Listening Proficiency by Age and Main Childhood Language (up to primary school age)
Table 13b: Listening Proficiency by Age and Main Childhood Language (from primary school age to school leaving age)
Table 14a: Listening Proficiency by Age and Main Language Used by Adults (up to primary school age)
Table 14b: Listening Proficiency by Age and Main Language Used by Adults (primary school age to school leaving age)
Table 15: Listening Proficiency by Age and Method of Acquisition
Table 16: Reading Proficiency by Age and Sex
Table 17: Reading Proficiency by Age and Region
Table 18: Writing Proficiency by Age and Sex
Table 19: Writing Proficiency by Age and Region
Table 20: Overall Proficiency by Age and Sex
Table 21: Overall Proficiency by Age and Region
Table 22: Overall Proficiency by Age and Satisfaction with Overall Proficiency
Table 23a: Relationship to Interlocutor by Degree of Active Domestic
Use and
Speaking Proficiency
Table 23b: Relationship to Interlocutor by Degree of Active Domestic Use and Age
Table 24a: Relationship to Interlocutor by Degree of Passive Domestic
Use and
Listening Proficiency
Table 24b: Relationship to Interlocutor by Degree of Passive Domestic Use and Age
Table 25a: Language Domain by Degree of Active Community Use and Speaking
Proficiency
Table 25b: Language Domain by Degree of Active Community Use and Age
Table 26a: Language Domain by Degree of Passive Community Use and Listening Proficiency
Table 26b: Language Domain by Degree of Passive Community Use and Age
Table 27a: Language Domain by Degree of Active Official Use and Speaking Proficiency
Table 27b: Language Domain by Degree of Active Official Use and Age
Table 28: Active Offer of Interpreter by Speaking Proficiency
Table 29a: Region by Access to Maori Radio and Listening Proficiency
Table 29b: Region by Listening to Māori Radio and Listening Proficiency
Table 29c: Age by Listening to Māori Radio and Listening Proficiency
Table 29d: Listening to Māori Language on Non-Māori Radio by Age and Listening Proficiency
Table 30a: Monthly Frequency by Age and Listening Proficiency
Table 30b: Hours per Week Listening to Māori Radio by Age and Listening Proficiency
Table 30c: Monthly Frequency by Age and Listening Proficiency
Table 31: How Much Māori Language Heard by Number of Hours per Week Listening to Māori Radio and Listening Proficiency
Table 32a: Satisfaction with the Amount of Māori Language Heard by Age and Listening Proficiency
Table 32b: Satisfaction with the Amount of Māori Language Heard by Number of Hours per Week Listening to Māori Radio and Listening Proficiency
Table 32c: Satisfaction with the Amount of Māori Language Heard by Age and Listening Proficiency
Table 33a: Age by Listening Proficiency and Watching Māori Television
Table 33b: Monthly Frequency by Age and Listening Proficiency
Table 33c: Satisfaction with the Amount of Māori Programming by Age and Listening Proficiency
Table 34a: Frequency of Reading in Māori by Age and Reading Proficiency
Table 34b: Frequency of Reading by Age and Reading Proficiency
Table 35: Type of Māori Language Material Read by Age and Reading Proficiency
Table 36a: Use of Dictionary if Unknown Word Encountered by Age and Reading Proficiency
Table 36b: Reading Proficiency by Dictionary Used if Unknown Word Encountered
Table 37a: Visibility of Māori Language in Official Contexts by Age and Frequency of Visibility
Table 37b: Satisfaction with Visibility of Māori Language in Official Contexts by Age and Overall Proficiency
Table 38a: Frequency in Writing in Māori by Age and Writing Proficiency
Table 38b: Frequency in Writing by Age and Writing Proficiency
Table 39: Type of Material Written by Age and Writing Proficiency
Table 40a: Participation in Māori Courses by Age and Overall Proficiency
Table 40b: Has been suppressed because it is subject to sample error too greatfor most practical purposes
Table 40c: Has been suppressed because it is subject to sample error too great for most practical purposes
Table 40d: Has been suppressed because it is subject to sample error too great for most practical purposes
Table 41a: Has been suppressed because it is subject to sample error too great for most practical purposes
Table 41b: Hours per Week (including study time) by Duration of Course
Table 42a: Type of Course by Impact on Māori Language Proficiency
Table 42b: Has been suppressed because it is subject to sample error too great for most practical purposes
Table 42c: Has been suppressed because it is subject to sample error too great for most practical purposes
Table 43a: Region by Overall Proficiency and Dependent Children Learning Māori
Table 43b: Has been suppressed because it is subject to sample error too great for most practical purposes
Table 43c: Venue for Child's Learning by Satisfaction with Child's Māori Language Learning
Table 44: Support for Māori Language Institutions by Age and Overall Proficiency
Table 45a1: Support for Māori Language Institutions (School Related) by Method of Support
Table 45a2: Support for Māori Language Institutions by Method of Support
Table 45b: Support for Māori Language Institutions by Monthly Support
Table 45c: Support for Māori Language Institutions by Hourly Support
Table 45d: Support for Māori Language Institutions by Receipt of Koha
Table 46: Use of Māori Language Material by Age and Overall Proficiency
Appendix Two: Survey Questionnaire and Showcards
This final report has been prepared for Te Puni Kōkiri, sponsors of the 2001 Survey on the Health of the Māori Language. The report contains information on the survey's objectives, methodology and classifications, as well as the set of tables that Statistics New Zealand agreed to provide Te Puni Kōkiri.
The 2001 Survey on the Health of the Māori Language was conducted as a post-census survey in May and June 2001. Respondents to the 2001 Census of Population and Dwellings, who identified with the Māori ethnic group, formed the base from which a sample of survey respondents were selected.
Almost 5,000 Māori were questioned about their ability to speak, listen, read and write in te reo Māori. Respondents were also asked about the environments in which they used the Māori language, and participation in Māori language education and revitalisation activities.
Survey results will be used to establish benchmarks for measuring Māori language revitalisation over time, and to provide the basis for language planning.
(Note: this final report replaces a provisional report released in December 2001.)
The purpose of the Survey on the Health of the Māori Language (MLS) is to describe the status of the Māori language by:
Knowledge of Māori language/proficiency
To estimate the number and proportion of Māori adults in the different speaking proficiency categories/listening proficiency categories/reading proficiency categories/writing proficiency categories.
To measure the extent to which Māori adults are satisfied with their overall level of speaking, listening, reading and writing proficiency.
Acquisition of Māori language
To describe the methods by which Māori adults acquired speaking proficiency by estimating how respondents learnt using various identified categories.
Use/visibility of Māori language
Speaking/hearing Māori
To measure the frequency with which Māori adults speak/hear Māori spoken in the home.
To measure the frequency with which Māori is spoken/heard in different community contexts.
To determine whether or not Māori has been spoken/heard in official government or government agency situations.
Reading/writing Māori
To measure the extent to which Māori adults have read/written in Māori (four week reference period).
To determine the types of Māori language material that Māori adults read by estimating the proportions in each reading-material-type category.
To identify the types of writing that Māori adults engage in.
To establish whether Māori adults use Māori dictionaries and which dictionaries they use.
Seeing Māori language material
To measure the perceived frequency with which official public material is seen in Māori and the satisfaction with the amount seen.
Language-oriented behaviour
Listening to/watching Māori radio & television
To measure the frequency with which Māori adults listen to Māori radio stations.
To investigate the Māori radio listening times of Māori adults.
To measure Māori adults satisfaction with the amount of Māori language on Māori radio stations.
To investigate why Māori adults don't listen to Māori radio stations.
To measure Māori adults' satisfaction with Māori language programming in mainstream radio/television.
To determine which Māori television programmes are being watched by Māori adults and the frequency with which they watch any of these programmes.
Current course attendance/last course attended
To estimate the proportion of Māori adults who are currently attending a Māori language course.
To identify what kind of courses are being undertaken by Māori adults, and the duration of those courses (within 12-month reference period).
To identify the fees and other types of costs associated with a Māori language course attended by Māori adults.
To investigate Māori adults' perceptions of the effectiveness (with reading, writing, listening and speaking) of their current/last course.
To measure the extent to which Māori adults are/were satisfied with their current course/last course.
To inquire into reasons for dissatisfaction with the course being attended/last course.
To estimate the proportion of Māori adults who are learning Māori informally.
Parent/caregiver to children under 15 years
To identify the characteristics of Māori adults who make education decisions for children, and the participation of those children in Māori language learning programmes.
To investigate the extent of satisfaction with children's Māori language learning programmes amongst Māori who make education decisions for children.
To inquire into reasons for dissatisfaction with children's language learning programme(s) amongst Māori who make education decisions for children.
To identify reasons why Māori who make education decisions for children don't chose Māori language learning programmes for children.
Participation in activities related to Māori language revitalisation
To measure the level of participation in Māori language related activities by estimating the proportion of Māori adults in each activity category.
To identify the form of involvement in Māori language revitalisation activities by estimating the proportions of Māori adults in each form of involvement category.
To estimate the number of hours that Māori adults spend in Māori language revitalisation activities.
Purchase of Māori language-related materials and resources
To determine the types of Māori language-related materials and resources purchased by Māori, by estimating the proportion of Māori adults in each material/resource-type category.
To measure the extent to which Māori adults use Māori language-related materials and resources by estimating the proportion of Māori adults in each material/resource-type category.
To estimate the amount spent on Māori language-related materials and resources.
To investigate the extent to which Māori demand for Māori language materials and resources is being met by examining Māori adults' perceptions of their availability.
The target population for the Survey on the Health of the Māori Language (MLS) is "Individuals belonging to the New Zealand Māori ethnic group who live in private households and who are aged 15 years and over".
The survey is restricted to Māori because of the practical difficulty and cost of surveying all New Zealanders while still producing good estimates for Māori. Also, the view taken by stakeholders consulted in the lead-up to the 1995 MLS was that, while the support of all New Zealanders is necessary, only Māori can ensure the survival of the language.
Māori is interpreted as being of Māori ethnicity, rather than Māori descent. A person is of Māori ethnicity if they specify 'Māori' as one of their ethnic groups in the 2001 Census.
Children are excluded because there are technical and administrative difficulties involved in surveying childhood language proficiency.
People living in non-private dwellings, eg institutions such as hospitals and prisons, or community-based dwellings such as university hostels and homes for the elderly are excluded from the target population. Note that although marae are classified as non-private dwellings, people who live permanently on marae are defined as being in a private dwelling.
The 2001 Census individual forms were used to develop the sample frame for the survey.
The survey population consists of individuals who returned individual census forms before the cut-off date for sampling and indicated that they were:
- of Māori ethnicity
- aged 15 years or over on census night (6 March 2001)
- in a private dwelling on census night
- not on an offshore island, except Waiheke Island, on census night.
Note that it is the person's census night address that determined their eligibility on the sample frame, not their usual residence address, because the physical forms were collected and stored by census night address. This means that people not at home on census night were selected on the basis of their census night address. Attempts were made to contact people who were no longer at their census night address when the survey was in the field.
Members of the target population excluded from the survey were:
Those for whom we did not receive a census form, or whose census form arrived after the sample was selected.
Visitors to non-private dwellings on census night who live in private dwellings, eg people staying in hotels, motor camps or hospitals on census night .
Those living in offshore islands, except Waiheke Island (these people are also excluded from most other Statistics New Zealand surveys because of the difficulty of surveying people in remote areas).
People who belong to the Māori ethnic group who did not answer the census ethnicity question.
Those initially selected in the MLS sample, but removed because they had also been recently selected in the Household Labour Force Survey (HLFS).
Out of scope people who may be wrongly included in the survey population are:
Occupants of non-private dwellings who happened to be staying at a private dwelling on census night.
Those whose census forms had ethnicity marked as Māori, but who did not identify as Māori during the MLS survey interview.
Survey respondents were asked to assess their own level of ability to speak, listen, read and write in te reo Māori. It was decided that self-assessment would be used to determine speaking, listening, reading and writing proficiency, following a literature review by Te Puni Kōkiri and a series of field tests which indicated that reasonable confidence could be attached to self-assessment scores.
Respondents were asked to place themselves into one of five categories as follows:
How well are you able to speak Māori in everyday conversation?
1 Very well (I can talk about almost anything in Māori)
2 Well (I can talk about many things in Māori)
3 Fairly well (I can talk about some things in Māori)
4 Not very well (I can only talk about simple/basic things in Māori)
5 No more than a few words or phrases.
How well are you able to understand spoken Māori?
1 Very well (I can understand almost anything said in Māori)
2 Well (I can understand many things said in Māori)
3 Fairly well (I can understand some things said in Māori)
4 Not very well (I can only understand simple/basic things said in Māori)
5 No more than a few words or phrases.
How well are you able to read Māori with understanding?
1 Very well (I can read almost anything in Māori)
2 Well (I can read many things in Māori)
3 Fairly well (I can read some things in Māori)
4 Not very well (I can only read simple/basic things in Māori)
5 No more than a few words or phrases.
How well are you able to write in Māori, with understanding?
1 Very well (I can write almost anything in Māori)
2 Well (I can write many things in Māori)
3 Fairly well (I can write some things in Māori)
4 Not very well (I can only write simple/basic things in Māori)
5 No more than a few words or phrases.
Respondents who rated their ability between 1 and 4 were then asked more detailed questions about the environments in which they heard or used te reo Māori. Those who rated their ability as 'no more than a few words or phrases' were not asked detailed questions about the environments in which they heard or used the Māori language.
Te Puni Kōkiri also specified a derived variable which combined the categories of speaking, hearing, reading and writing into an overall proficiency rating with output categories of:
Very high [proficiency]
High [proficiency]
Medium [proficiency]
Low [proficiency]
None [proficiency]
The 2001 Survey on the Health of the Māori Language (MLS) was run as a post-census survey. A post-census survey provides a unique opportunity to run a large survey of a small sub-group of the population in a cost-effective manner. Administrative lists such as the electoral roll suffer from serious undercoverage, and using the Statistics New Zealand household survey frame is expensive because of the need to screen a large number of households in order to find the target population. The 2001 Census frame provides a degree of coverage not matched by any other single method.
In addition, information collected from respondents in the MLS can be linked with their census data on variables such as income and labour force status. This minimises respondent burden and data collection costs.
New Zealand was divided into 407 census districts in 2001. These were further subdivided into 5,800 sub-districts and 37,000 meshblocks. Sample selection for the MLS took place in three stages. First 119 districts were selected. Within these districts, meshblocks and sub-districts were selected. Within selected meshblocks a sample of people aged 15-54 years was selected and within sub-districts a sample of people aged 55 years and over. The districts were selected over eight regions to achieve a balance over the country. People were selected in two age groups to ensure that enough people aged 55 years and over were in the sample.
The selection of geographic areas was with probability proportional to size, rather than the more usual simple random sample. This results in a more efficient sample when a population is not spread evenly over areas as is the case here the Māori population is quite heavily concentrated in some areas while many areas have small numbers of Māori.
The selection of people was done manually in the district offices where staff selected a systematic sample by going through census forms according to instructions. A sample of 6,400 people was selected.
Two types of error are possible in an estimate based on a sample survey: sampling error and non-sampling error. Sampling error is a measure of the variability that occurs by chance because a sample, rather than the entire population is surveyed. The size of the sample error has been controlled by the size of the sample, the use of probability proportional to size selection, stratification, sample allocation and estimation techniques.
There may also be inaccuracies that occur because of imperfections in reporting by interviewers and respondents and errors made in the coding and processing of data. These are referred to as non-sampling error. Statistics New Zealand endeavours to minimise the impact of these errors through the application of best survey practices. It is not possible to quantify the non-sampling error.
Sample errors are given for some key results in Tables 1a, 1b, 2a and 2b. For example, the estimated total number of Māori adults who speak Māori 'fairly well' is 34,700. This figure has a sampling error measured at the 95 percent confidence level of plus or minus 3,400 (or 10 percent). This means that there is a 95 percent likelihood that the true value lies between 31,300 and 38,100.
However, space does not allow for the separate indication of sample errors for all estimates. Sample errors and relative sample errors (RSEs) for various size estimates are given in the tables below. RSEs are the sample error as a percentage of the estimate. These figures will not give a precise measure of the sample error for a particular estimate, but will provide an indication of its magnitude.
As the sample errors in the tables below show, the smaller the estimate, the higher the RSE. Very small estimates are subject to such high sample errors (relative to the estimate) as to detract seriously from their value for most reasonable uses. In the tables provided with this report, only estimates with RSEs less than 30 percent are considered sufficiently reliable for most purposes. However, estimates with larger RSEs, between 30 percent and less than 50 percent, have been included and are preceded by an asterisk to show that they are subject to high sampling error and should be used with caution. Estimates with RSEs of 50 percent or more are considered unreliable for most uses - the data in these cells has been replaced by a double asterisk.
Table of relative sample errors for person
estimates
in the age group 55 years and over
|
Size of Estimate |
Sample Error |
(a) Relative Sample Error (%) |
|
|
(b) |
200 |
200 |
100 |
|
(b) |
300 |
205 |
68 |
|
(b) |
400 |
230 |
57 |
|
(b) |
500 |
250 |
50 |
|
(c) |
750 |
290 |
39 |
|
(c) |
1000 |
350 |
35 |
|
(c) |
1500 |
455 |
30 |
|
2000 |
540 |
27 |
|
|
3000 |
650 |
22 |
|
|
4000 |
730 |
18 |
|
|
5000 |
775 |
15 |
|
|
7500 |
790 |
10 |
|
|
10000 |
1040 |
10 |
|
|
15000 |
1190 |
8 |
|
|
20000 |
1240 |
6 |
|
|
30000 |
1135 |
4 |
Table of relative sample errors for person
estimates
except age 55 and over
|
Size of Estimate |
Sample Error |
(a) Relative Sample Error (%) |
|||
|
(b) |
300 |
300 |
100 |
||
|
(b) |
400 |
350 |
87 |
||
|
(b) |
500 |
390 |
77 |
||
|
(b) |
750 |
420 |
56 |
||
|
(b) |
1000 |
500 |
50 |
||
|
(c) |
2000 |
760 |
38 |
||
|
(c) |
3000 |
940 |
31 |
||
|
4000 |
1070 |
27 |
|||
|
5000 |
1160 |
23 |
|||
|
6000 |
1210 |
20 |
|||
|
7000 |
1240 |
18 |
|||
|
10000 |
1490 |
15 |
|||
|
15000 |
1920 |
13 |
|||
|
20000 |
2260 |
11 |
|||
|
30000 |
2760 |
9 |
|||
|
40000 |
3080 |
8 |
|||
|
50000 |
3270 |
7 |
|||
|
75000 |
3320 |
5 |
|||
|
100000 |
2920 |
3 |
|||
|
150000 |
1210 |
1 |
|||
(a) Relative sample errors (RSE) are the sample error as a percentage of the estimate.
(b) Estimates with an RSE of 50 percent or more. These estimates are considered unreliable for most purposes.
(c) Estimates with an RSE between 30 percent and less than 50 percent. These estimates should be treated with caution.
A total of 6,092 eligible people were selected for the survey. The response rate was 78 percent, with 4,738 full responses received. A further 300 people were selected but were found in the field to be out of scope, mainly because they did not have Māori ethnicity.
The main reason for non-response was non-contact (13 percent of those eligible), with around half non-contacts having moved from their address reported at the census. For those contacted, the refusal rate was 7 percent and a further 4 percent either gave partial answers or were unable to respond due to death or illness.
The survey was completed by personal interview throughout the months of May and June 2001. Respondents were offered the choice of completing the survey in either te reo Māori or English, with the layout of the questionnaire allowing respondents to switch from one to the other if necessary. Some 550 interviews were completed in either te reo Māori or a combination of te reo Māori and English, representing 12 percent of the total number of interviews.
Bilingual prenotification letters were sent out two weeks before interviewing began. Although addresses were sometimes insufficient for delivery, letters were appreciated by respondents who received them. Good use was made of an 0800 number available to respondents and interviewers. Support for interviewers extended to evenings and weekends.
Interviews in te reo Māori took on average two hours to complete, while those in English were closer to 40 minutes. Respondents who had identified on their census forms that they could hold a conversation about everyday things in Māori were assigned a fluent interviewer. In other cases, if a respondent chose to complete the interview in te reo Māori, they were referred to a fluent speaker.
A Māori liaison officer was appointed to take overall responsibility for recruitment of the survey interviewers. Sixty-three interviewers worked on the survey, and fifty of these were Māori. Forty-six interviewers were fluent speakers, and ten of these had recently worked as Kaitakawaenga for the 2001 Census. Twelve interviewers were existing Statistics New Zealand interviewers.
All interviewers working on the survey had a good knowledge of the communities they were working in and this proved invaluable in encouraging participation.
The fieldforce received two days of training. This was marae based, in three locations; Wellington, Rotorua and Auckland. The training included role plays of the questionnaire, field procedures and stressed the kaupapa of the survey. The Chair of Statistics New Zealand's Māori Statistics Forum acted as facilitator for the training sessions and Te Puni Kōkiri presented the objectives of the survey.