Occupation (information about this variable and its quality)

Priority level

Priority level 3

We assign a priority level to all census variables: Priority 1, 2, or 3 (with 1 being highest and 3 being the lowest priority).

Occupation is a priority 3 variable. Priority 3 variables do not fit in directly with the main purpose of a census but are still important to certain groups. These variables are given third priority in terms of quality, time, and resources across all phases of a census.

The census priority level for occupation remains the same as 2013.

Quality Management Strategy and the Information by variable for occupation (2013) have more information on the priority rating.

Overall quality rating for 2018 Census

Moderate quality

Data quality processes section below has more detail on the rating for this variable.

The External Data Quality Panel has provided an independent assessment of the quality of this variable and has rated it as poor quality. 2018 Census External Data Quality Panel: Assessment of Variables has more information.

Caution is advised when using this variable at small geographies. Please see Recommendations for use and further information section below.

Subject population

Employed census usually resident population aged 15 years and over

‘Subject population’ means the people, families, households, or dwellings to whom the variable applies.

How this data is classified

Australian and New Zealand Standard Classification of Occupations (ANZSCO V1.2.0)

ANZSCO is structured with five hierarchical levels:

1. Major group (9 categories with 1 digit codes)

2. Sub-major group (44 categories with 2 digit codes)

3. Minor group (99 categories with 3 digit codes)

4. Unit group (361 categories with 4 digit codes)

5. Occupation (1,035 categories with 6 digit codes)

ANZSCO is a skill-based classification used to classify all occupations and jobs in the Australian and New Zealand labour markets.

The conceptual model adopted for ANZSCO uses a combination of skill level and skill specialisation as criteria to design major groups which are meaningful and useful for most purposes. The eight major groups are formed by grouping together sub-major groups using aspects of both skill level and skill specialisation.

The Standards and Classifications page provides background information on classifications and standards.

The skill level criterion is applied as rigorously as possible at the second level of the classification, the sub-major group level, together with a finer application of skill specialisation than that applied at the major group level.

The major groups are:

1 Managers

2 Professionals

3 Technicians and Trades Workers

4 Community and Personal Service Workers

5 Clerical and Administrative Workers

6 Sales Workers

7 Machinery Operators and Drivers

8 Labourers

9 Residual Categories (Operational Codes only)

'Residual Categories' includes response unidentifiable, response outside scope, and not stated.

For the 2018 Census, we only coded occupation responses to ANZSCO. In 2013, we dual-classified occupation data to both ANZSCO and the previous occupation classification, NZSCO99.

Since 2013, ANZSCO has had minor changes at the lowest level. These include updating occupation names, and often codes, as occupations become more specific. For example, ‘ship’s surveyor’ was updated to ‘marine surveyor’ with the same code of ‘231215’. A more detailed occupation in the latest classification is ‘Registered Nurse (Paediatric)’ (254425) which used to be coded to the catch-all category of ‘Registered Nurse Not Elsewhere Classified’ (254499).

These alterations have not fundamentally changed the classification, meaning that 2018 Census data can still be compared to 2006 and 2013 Census data (except for those specific occupations). No concordance is necessary.

Question format

Occupation data was collected on the individual form (question 40 on the paper form). Stats NZ Store House has samples for both the individual and dwelling paper forms.

There were differences between paper forms and online forms.

Online:

• we introduced an ‘as-you-type’ list, which is a drop-down menu of occupations selected from a list of probable survey responses and their classification categories
• this list reduced vague text responses online, which reduced the need for manual coding.

Paper:

• respondents wrote in their occupation, which can lead to vague responses for example manager, nurse, teacher
• paper forms required scanning. Where answers could not be automatically coded, they required manual coding.

Data from the online forms may be better quality than data from paper forms due to the improvement in coding.

There were also differences in how occupation was collected between censuses.

In 2018:

• we removed the supplementary question on ‘tasks and duties’ in the job, which was asked in previous censuses.

In 2013:

• we used responses to the ‘tasks and duties’ question during manual intervention as additional information to help manual operators code occupation responses.

How this data is used

Outside Stats NZ

• Monitoring changes in occupational concentrations, developing curriculum training policies, and international comparisons.
• Determining the characteristics of people within selected occupational groups.
• Analysing structural changes in the labour market over time, the study of occupational accidents, mortality, and morbidity rates.
• Analysing and classifying socioeconomic status in studies of social disadvantage, poverty, and equity.

Within Stats NZ

• Analysing and monitoring structural changes in the labour market, and planning for new demand in occupation resulting from technological or economic changes.

2018 data sources

We used alternative data sources for missing census responses and responses that could not be classified or did not provide the type of information asked for. Where possible, we used responses from the 2013 Census, administrative data from the Integrated Data Infrastructure (IDI), or imputation.

The table below shows the breakdown of the various data sources used for this variable.

Please note that when examining occupation data for specific population groups within the subject population, the percentage that is from statistical imputation may differ from that for the overall subject population.

Missing and residual responses

In 2018, missing occupation responses were replaced by data derived by statistical imputation. In previous censuses, we did not impute occupation and missing responses were coded to ‘not stated’.

Percentage of ‘not stated’ for the census usually resident employed population aged 15 years and over:

• 2018: 0.0 percent
• 2013: 2.8 percent
• 2006: 3.8 percent.

In 2018, there were no other residual responses remaining in the data after imputation.

In previous censuses, responses that could not be classified or did not provide the type of information asked for such as ‘response unidentifiable’ remain in the data where we have been unable to find information from another source. For output purposes, these residual category responses were grouped with ‘not stated’ and classified as ‘not elsewhere included’.

Percentage of ‘not elsewhere included’ for the census usually resident employed population aged 15 years and over:

• 2018: 0.0 percent
• 2013: 5.1 percent
• 2006: 5.7 percent.

2013 Census data user guide provides more information about non-response in the 2013 Census.

Data quality processes

Overall quality rating: Moderate quality

Data was evaluated to assess whether it meets quality standards and is suitable for use.

Three quality metrics contributed to the overall quality rating:

• data sources and coverage
• consistency and coherence
• data quality.

The lowest rated metric determines the overall quality rating.

Data quality assurance for 2018 Census provides more information on the quality rating scale.

Data sources and coverage: Moderate quality

We have assessed the quality of all the data sources that contribute to the output for the variable. To calculate a data sources and coverage quality score for a variable, each data source is rated and multiplied by the proportion it contributes to the total output.

The rating for a valid census response is defined as 1.00. Ratings for other sources are the best estimates available of their quality relative to a census response. Each source that contributes to the output for that variable is then multiplied by the proportion it contributes to the total output. The total score then determines the metric rating according to the following range:

• 98–100 = very high
• 95–<98 = high
• 90–<95 = moderate
• 75–<90 = poor
• <75 = very poor.

Data sourced through statistical imputation was moderately comparable to census forms. The proportion of occupation data sourced from statistical imputation contributed to the score of 0.90, determining the moderate quality rating.

Data sources, editing, and imputation in the 2018 Census has more information on the Canadian census edit and imputation system (CANCEIS) that was used to derive donor responses.

Consistency and coherence: High quality

Occupation data is consistent with expectations across nearly all consistency checks, with some minor variation from expectations that makes sense due to real-world change, incorporation of other sources of data, or a change in how the variable has been collected.

There are only minor comparability issues for certain occupations when comparing with previous years, due to changes at the lowest level of the classification and improved coding.

We note that, even with some inconsistencies and minor changes to the classification, the 2018 data can still be compared with previous censuses.

Data quality: High quality

There are only minor data quality issues for occupation. The quality of coding and responses within classification categories is high and the distribution of occupations meets expectations.

The use of as-you-type lists for online forms, in conjunction with a high online response, generally resulted in good quality data from received responses due to the reduction in vague responses.

Recommendations for use and further information

We recommend that the use of the data can be similar to that produced in 2013.

However, when using this data you should be aware that:

• caution is advised when using this variable and when cross-tabulating with workplace address variable at small geographies. At small geographies, there will be variability in the percentage of imputation for a given area. This means some small geography areas will have poorer quality data than the overall quality rating.
• the use of statistical imputation means there is no non-response category for 2018
• the use of imputation may have slightly inflated the total number of respondents in all categories. Care should therefore be taken if comparing absolute figures to previous years. We recommend using proportions, particularly at the lowest level of the classification, where there may be inconsistencies with the time series due to the use of an as-you-type list in 2018.

Contact our Information Centre for further information about using this variable.