Methodology

Job openings: 

Lightcast applies the cyber security definition across its proprietary database of job postings to identify the total number of cyber security job postings within a given geography, occupation, and industry over the specified time period.

Workers: 

Lightcast uses both Australian Bureau of Statistics (ABS) employment data and its proprietary job posting data to compute employment statistics. To derive employment estimates across geographic regions within Australia, the totals across each sex from Table RQ2, “Employed persons by Labour market region (ASGS), Occupation major group (ANZSCO) and Sex, Annual averages of the preceding four quarters, Year to August 1999 onwards ” are weighted by the percentage of historical cyber security job postings within each geographic region and ANZSCO Occupation major group. This methodology is used to produce more granular employment estimates than what is available in the public ABS dataset.

Worker gender: 

To provide estimates of the gender breakdown within each geography, occupation and industry, Lightcast utilise the employment values calculated via the methodology that is described above and then apply gender weights from ABS Table EQ08, “Employed persons by Occupation unit group of main job (ANZSCO), Sex, State and Territory, August 1986 onwards”. Weights from this table are specific to an ANZSCO Occupation unit group and geography (national and state level). Weights are consistent across industries, and SA4-level weights are derived from the state-level data. The weights used for the aggregate cyber security ANZSCO classification (All other cyber roles) are derived from the averages across the broader ICT industry.

In order to ensure that the gender breakdowns sum up appropriately across ANZSCO occupations and geographic regions, Lightcast performed slight redistributions of the worker gender estimates produced using the ABS weights. Consequently, the final weights vary slightly from what is produced by ABS in Table EQ08.

All other cyber roles ANZSCO classification: 

This classification includes roles which require cyber security skills, certifications, or expertise, but are not classified under the cyber related ANZSCO Unit Group codes. These roles are commonly classified under other ICT related ANZSCO Unit Group codes, such as “2631 Computer Network Professionals”.

City: 

City data on the website reflects all of the SA4 regions within Australia with the exception of the following locations: Central Queensland, Australian Capital Territory (this data is provided at the state level), Other Territories, Migratory - Offshore - Shipping locations, and No usual address locations. Consequently city-level data is not meant to sum up to state or national-level totals.

Regional & Urban: 

Roles advertised without a location, i.e. online roles, have been redistributed between regional & urban.

NICE Work Roles: 

To calculate the NICE work role category distribution, Lightcast reviewed the tasks and KSAs associated with each work role and mapped them to our closest corresponding skills, job titles, and certifications, customising one or more matching NICE work roles for each job posting.

Geographic Concentration: 

The geographic concentration is calculated by comparing the concentration of cyber security job postings to other job postings—and comparing that to the national average.

Certification Holders: 

Certification holder data is aggregated directly from data provided by the certifying organisation. Values of less than 10 are suppressed to protect the privacy of these individual certification holders.

Certifications requested: 

To classify the frequency with which certifications are demanded, Lightcast looks for exact matches of the certification name within the job posting. Inferring certifications based on acronyms is limited due to the fact that there can be one acronym that refers to multiple different certifications.

Demand: 

Lightcast applies the previously described cyber security definition across its proprietary database of job postings to identify the total number of cyber security job postings within a given country during the specified year.

Employment: 

To produce cyber security employment estimates across different countries, Lightcast uses the same methodology that is used within Australia. Data for the United Kingdom comes from the Office for National Statistics (ONS), while United States data comes from the Bureau of Labor Statistics (BLS).

Salary: 

The salaries are the average, annual values of advertised job postings. Salaries are aggregated across full-time and contracted positions, with the salaries for contractors being converted to annual averages.

The Career Pathways data provides insights into career progression and transition pathways, and is constructed from the core jobs, skills and certifications advertised in the cyber security landscape. Career pathways are constructed on both the role and certification level using Lightcast’s proprietary database of job postings and online professional social profiles.

Aggregate statistics are provided for the ‘top’ cyber security roles across average salary, job openings, requested education, common job titles, top skills and certifications requested, from 1st February 2022 to 31st January 2023. As this data is provided only for roles identified as ‘top roles’, it will not be equivalent to the data presented on the geographic supply & demand maps on the site.

The roles are classified based on Lightcast’s own Occupation Taxonomy, which has been mapped to the National Initiative for Cybersecurity Education (NICE) Framework. Custom mappings between the Lightcast Occupation Taxonomy and the ANZSCO taxonomy were created by Lightcast in consultation with AustCyber.

Supply and Demand: 

To forecast growth in cyber security supply and demand, Lightcast trained multiple Autoregressive Integrated Moving Average (ARIMA) models using historical supply and demand data. (Note that all historical data sets were created using the methodologies outlined in the job openings and worker sections of “Current Supply and Demand Data”). An individual ARIMA model was trained for each combination of geographies, ANZSCO unit groups and ANZSIC industries. Akaike information criterion (AIC) and Bayesian information criterion (BIC) was used to determine the optimal order for the ARIMA model parameters. Training individual models rather than one for the aggregate of cyber security supply and demand data allowed Lightcast to capture nuances in the trends within certain locations, occupations or industries. The 2022 worker supply and job posting figures were used as the base year in the model, to output the yearly growth rates between 2023 and 2027. This output was then benchmarked against the employment projections produced by the National Skills Commission (NSC). Employment growth rates from the ARIMA model were capped based on the maximum growth rate provided by the NSC for ICT Occupations. Growth rates were then applied to the current values to estimate employment and job postings for each year from 2023 to 2027.