2975.0.55.016 - Census Working Paper 96/4 - Fact sheet 16 - Labour Force Status, 1996  
ARCHIVED ISSUE Released at 11:30 AM (CANBERRA TIME) 15/09/1999   
   Page tools: Print Print Page Print all pages in this productPrint All

1996 CENSUS OF POPULATION AND HOUSING
FACT SHEET 16
LABOUR FORCE STATUS


This Fact Sheet addresses form design and coding changes to Labour Force Status that occurred between the 1991 and 1996 Censuses and their subsequent impact on labour force figures. These changes impact on the comparability of 1991 and 1996 Labour Force Status, and care should be taken when working with these data.

Labour Force Status is applicable for all persons 15 years and over. The Census uses a series of questions to determine whether persons are employed, unemployed, or not in the labour force. Employed people are categorised by Status in Employment. Unemployed people are classed as looking for full-time or part-time work.


CHANGES TO QUESTION STRUCTURE

1. A new question, Availability to Start Work, was included in the 1996 Census to improve comparability of Labour Force data (particularly unemployment data) between the Census and the Labour Force Survey. The inclusion, in 1996, of the question on Availability to Start Work resulted in some people who would previously have been classed as Unemployed, being classed as Not in the Labour Force (NILF).

2. The order of the Labour Force questions was changed in 1996 to reduce the non-response problems, encountered in 1991, associated with people incorrectly following sequencing instructions. This change in sequence also facilitated the inclusion of the Availability to Start Work question. The new order involved moving the Hours Worked and Looked for Work questions to the end of the employment related questions later in the series of Labour Force questions.

3. Additional categories were included in the Job Last Week question to differentiate between self-employed people in limited liability companies and those not in limited liability companies. This was done to improve comparability between the Census and Labour Force Survey in the measurement of Wage and salary earners.

4. The Labour Force Status output codes used in 1991 were reworded slightly for 1996 in accordance with the relevant standard. Wage and Salary Earner was changed to Employee, Self Employed to Own Account Worker and Unpaid Helper to Contributing Family Worker.


EFFECTS ON DATA

a. Status in Employment

1. The difference between the 1991 Census data and 1996 Census data, for Status in Employment, is largely due to the introduction of the limited liability categories as per Question 31 on the 1996 Census form.

2. In 1991, compared to the Labour Force Survey, the Census overstated the number of employers and self employed and understated the number of employees. The limited liability response categories were introduced to correct this. However, rather than just bringing the Census into line with the Labour Force Survey, the introduction of the limited liability categories seems to have reversed the relationship between the two sets of data.

3. Compared to the Labour Force Survey, the 1996 Census now overstates the number of employees and understates the number of employers and self-employed. More importantly for users of Census labour force data, movements between 1991 and 1996 Censuses have been distorted.

TOTAL EMPLOYED, AUSTRALIA

    1991 & 1996 Censuses
1991 Census
1996 Census
% change between counts
Difference
96% - 91%

Count
%
Count
%
    Employee
5,807,575
81.69
6,921,395
90.64
19.18
8.95
    Employer
485,727
6.83
194,622
2.55
-59.93
-4.28
    Own account worker
753,489
10.60
442,386
5.79
-41.29
-4.81
    Contributing family worker
62,545
0.88
77,905
1.02
24.56
0.14
    Total Employed
7,109,336
100.00
7,636,308
100.00
7.41


4. However, when allowance is made for the impact of the Limited Liability categories on 1996 data, the intercensal movements are much more reasonable.

TOTAL EMPLOYED, AUSTRALIA

    1991 & 1996 Censuses
    (1996 Adjusted for Ltd Liability)
1991 Census
1996 Census Adjusted for Ltd Liability
% change between counts
Difference
Adj% - 91%

Count
%
Count
%
    Employee
5,807,575
81.69
6,357,279
83.25
9.47
1.56
    Employer
485,727
6.83
502,145
6.58
3.38
-0.26
    Own account worker
753,489
10.60
698,979
9.15
-7.23
-1.45
    Contributing family worker
62,545
0.88
77,905
1.02
24.56
0.14
    Total Employed
7,109,336
100.00
7,636,308
100.00
7.41


5. It is not possible to know whether the adjustments shown in the above table offset all of the impact of the form design changes on comparability of Status in Employment data. Accordingly, users should be cautious in interpreting the apparent shifts between categories.

b. Industry for Employed Persons

1. The non-response rates for Industry and Industry Sector decreased substantially - for Industry, from 6.86% in 1991 to 1.61% in 1996. This could reflect both coding and form design changes.

  • Coding: The business coding methodology was changed to improve the match rate to the business register and a procedure was introduced to query any not stated Industry responses when an occupation or business name was stated.

  • Form design: The order of the labour force questions was changed to avoid the problems with sequencing instructions encountered in 1991. It should also be noted that the position of the Industry question on the form moved from the bottom of the form in 1991 to near the top of the form in 1996. Changes to the position of questions on the form are known to affect non-response rates.
c. Edits in Labour Force Status

1. The 1996 edits were largely unchanged from 1991. The exception to this involved the edit where Females aged greater than 60 and Males aged greater than 65 with a derived Labour Force Status of Not Stated were automatically edited to the 'Not in Labour Force'. This edit was applied more strictly in 1996 than it had been in 1991. The effect of this is minor, with more people being set to Not in the Labour Force rather than to Not Stated.