Firstly, I derived the Average Seasonal Job Vacancies (2014–2010) and the % Change in Vacancies graph by obtaining data from the ABS Job Vacancies Survey (JVS) spanning 4 years with the notable exception of 2015 — the year (as of 12/01/2016) was incomplete. I averaged the seasonally unadjusted original data of the present quarter to its last quarter over the 4 years. Taking for example, the change in vacancies of the August to November quarter in 2014 (2.7%) is added to the previous year (2013) and so on, which is then averaged over the number of years. Likewise, this was also done regarding the percentage change in vacancies provided in the ABS report.

I originally compiled the data from the JVS Surveys spanning 2000–2014 (Figure 1a & 1b in the appendix) to ensure that the trends were consistent and reflective of a longer period. However, compiling data from a number of years included the volatile fluctuations in the data from the early 2000s. For example, the % change in vacancies - in August 2002, the % change in the original seasonally unadjusted data is 21.4 which compared to a similar time span (in August 2014 — is 4.1) is much greater.

Furthermore, the statistics used in one publication differed greatly from the same statistic used in a later publication. In the May 2002 publication of the job vacancy report, May 2001 is placed at 83.4 seasonally unadjusted for Australia. This differs vastly from the August 2001 publication where it is 87.4. This discrepancy is accounted for by the ABS where the “sampling variability (is) too high for most practical purposes.” Thus, this data was difficult to analyse due to fluctuations and the enormity of the data.

From this, I sought to construct the graphs from 2005 to 2014 (Figure 2a & 2b). Here, this data is limited because of the statistical changes introduced and implemented in 2009, as well as the period when the JVS was suspended, producing a gap in the series (5 periods between August 2008 and August 2009 inclusive are not accounted for). Through this process, I tried to avoid incomplete years so that the data would be in essence more ‘complete’. Moreover, in ensuring that my information was consistent, I decided to use the data from the implementation onwards and sought not to compare estimates from November 2009 onwards with estimates from May 2008 and the previous periods.

Both of these time periods have been discounted due to the reasons I have listed above. However, they are available in the appendixes for further interest.

The process for producing the average participation rate of the labour force graphs was definitely easier compared to the vacancy graphs. The data was compiled monthly by the ABS, which allowed for the trends to be seen and analysed more efficiently. Furthermore, the end result matched most of the research I had conducted as well as validated by several of my hypotheses.

Whilst drawing up these graphs, certain trends were less evident in the Australian data as a result of the vacancy data being compiled quarterly; this frustrated me. However, the main reason of this exercise in graphing was validating hypotheses and thus its purpose has been achieved. Warmer months have more job vacancies and the Christmas season is a big time for recruitment. University holidays also signal a time for recruitment to begin whilst graduation and internship recruitment cycles bolster the statistics.

Appendix

For reference, the most present graph of the trend for overall job vacancies:

Perhaps of further interest,