Page tools: Print Page Print All | ||||
CHAPTER 27. JOB VACANCIES SURVEY
27.4 Data published for the job vacancies series by sector are available on an original, seasonally adjusted, and trend basis. Industry and state data are only published on an original basis. As a result of JVS being suspended in 2008 and 2009, there is a gap in all series: original, seasonally adjusted and trend, for five quarters between August 2008 and August 2009 inclusive. 27.5 Estimates are compiled according to the concepts and definitions outlined in Chapter 4: Employment measures and classifications (paid employees) and Chapter 11: Job vacancies. SCOPE 27.6 The scope of the survey is restricted to employing businesses. In addition, the standard scope exclusions for ABS labour-related business surveys (outlined in Chapter 25: Methods used in ABS business surveys) apply to this survey. SURVEY METHODOLOGY 27.7 Details of the total number of job vacancies available for immediate filling on the survey reference date are obtained on a quarterly basis from selected businesses. Data is collected using Computer Assisted Telephone Interviewing (CATI), whereby an interviewer enters the data into a computer during a telephone interview. 27.8 The survey reference date for job vacancies is the third Friday in the middle month of the quarter. 27.9 Follow-up procedures are in place to obtain information from respondents who are unable to provide data at the time of the initial interview. A minimum response rate for the survey of 95% across states/territories and industry divisions is usually achieved each quarter. SAMPLE DESIGN 27.10 A probability sample of statistical units (employing businesses) is drawn from the ABS Business Register using the process outlined in Chapter 25: Methods used in ABS business surveys. Variables used to stratify the survey frame are:
27.11 Statistical units with benchmark employment greater than a set cutoff (this cutoff will vary for different states/territories) are completely enumerated. Strata with a very small number of statistical units may also be completely enumerated, but such strata may become sampled strata if the number of units increases sufficiently. 27.12 In addition to constraints outlined in Chapter 25: Methods used in ABS business surveys, sample selection is constrained by ensuring that there is minimum overlap with other labour-related business surveys. SAMPLE SIZE AND ALLOCATION 27.13 Approximately 5,000 statistical units are selected in the sample to yield a live sample of approximately 4,800 units. 27.14 The sample is allocated optimally across the strata using a technique designed to minimise the variance of job vacancies estimates at both the national and state/territory level. SAMPLE ROTATION 27.15 The sample is updated each quarter to reflect changes in the ABS Business Register. Approximately 8% of the sample for the non-completely enumerated strata is replaced each quarter. The JVS population is updated quarterly to take account of:
27.16 Sample rotation is implemented for the majority of strata, but is not implemented where the population of a stratum is so small that units rotating out of the sample would be rotated back in after only a short interval. ESTIMATION 27.17 Number raised estimation is used in all strata. 27.18 For non-responding units in the sampled strata, the Live Respondent Mean method of imputation is used. 27.19 For non-responding units in the completely enumerated (CE) strata, an imputed growth rate is applied to the most recent reported data for the unit, provided that data have been reported in the previous quarter. Growth rates are estimated for each industry division based on data provided by CE units for the most recent quarter provided. Where data for non-responding CE units have not been reported in the previous quarter, ratio imputation is used. The ratio of job vacancies to benchmark employment is calculated at industry division level for responding units from the current quarter. This ratio is then applied to the benchmark employment for the non-responding unit to arrive at the imputed value for job vacancies. 27.20 Survey outliers are treated using the 'surprise outlier' technique. 27.21 Survey estimates include an adjustment called Business Provisions to allow for births and deaths of businesses that have occurred up to the end of the survey reference period but which are not reflected on the survey frame. 27.22 For further information on estimation methods used in ABS Business Surveys, refer to Chapter 25: Methods used in ABS business surveys. RELIABILITY OF THE ESTIMATES 27.23 Estimates from the survey are subject to both sampling and non-sampling error. The 27.24 The 'jack-knife' approach is used to calculate estimates of variance for this survey. For further information on the jack-knife technique for calculating variance, or on sampling and non-sampling error, refer to Chapter 17: Overview of survey methods. DATA COMPARABILITY OVER TIME 27.25 The JVS was suspended following the May 2008 survey and was reinstated for the November 2009 survey. As a result of JVS being suspended, there is a gap in all series: original, seasonally adjusted and trend, for five quarters between August 2008 and August 2009 inclusive. The ABS cannot produce reliable estimates by collecting this missing data retrospectively, and has not been able to fill the gap using other data sources. For further information, see the Information Paper: Reinstatement of Job Vacancies Survey (cat. no. 6354.0.55.001). 27.26 In order to provide a high degree of consistency and comparability over time, changes to survey methods, concepts, data item definitions, and frequency of collection are made as infrequently as possible. Significant changes have included:
Annual Job Vacancies Survey via mail-out commenced (largely to investigate practicability of a JVS). 1977 Introduction of a smaller scale quarterly telephone-based survey. Sample based on lists of private and public employers. 1978 Annual and quarterly surveys discontinued. 1979 Quarterly survey reintroduced. Treatment of Australian Public Service vacancies changed to exclude "vacancies" only available to public service employees. 1980 First collection of job vacancies registered with the Commonwealth Employment Service (CES) (continued on annual basis). 1982 Collection of vacancies classified by sex discontinued. 1985 Job vacancies data published by sector for the first time. 1988 ABS publication of job vacancies registered with the CES discontinued. This data was available via special data service in 1988. 1989 Seasonally adjusted series produced for the first time (November). Collection of job vacancies registered with CES discontinued. Job Vacancies, Australia (cat. no. 6231.0) and Overtime, Australia (cat. no. 6330.0) merged into Job Vacancies and Overtime, Australia publication (cat. no. 6354.0). 1993 Trend estimates published for the first time. 1994 Survey redesigned on an ANZSIC (1993) industry basis. The historical Industry series data was backcast on an ANZSIC 1993 basis. Sample rotation increased from approximately 5% to approximately 8% in rotating strata. 1998 Treatment of Australian Public Service vacancies changed (from being excluded to being included) after vacancies were made available to all Australian citizens. 1999 Introduction of Live Respondent Mean imputation for the sampled sector, and the Business Provisions adjustment for the private sector. Overtime component discontinued. Significant improvement in procedures, particularly coverage of vacancies within statistical units. 2002 Changes to the ABS Business Register and the ABS statistical units model arising from New Tax System. Changes did not affect the continuity of the key statistical series. 2003 Collection of number of employees discontinued. Publication of job vacancy rate discontinued. 2006 Concurrent seasonal adjustment method introduced, replacing the forward factor adjustment method previously used. 2008 Survey suspended for five quarters from August 2008 to August 2009 inclusive. 2009 Survey reinstated for the November 2009 reference period, with a new sample based on ANZSIC 2006. Improvements to survey frame meant that sample was reduced from 5,200 to 5,000 units with no reduction in survey accuracy. Survey redesigned on ANZSIC 2006 basis from November 2009, but historical ANZSIC 1993 series up to May 2008 not backcast. 2010 Estimates from reinstated survey first published after the February 2010 reference period. Seasonally adjusted and trend estimates for November 2009 onwards reintroduced from the August 2010 issue. FURTHER INFORMATION 27.27 For further details contact the Labour Market Statistics Section, on Canberra (02) 6252 7206 or email <labour.statistics@abs.gov.au>.
|