Important information in interpreting forecast ranges

In interpreting forecast ranges, Demand Map users should note:

Forecast certainty

• These ranges are forecasts for the scheme once it is fully developed (expected to be by 2023). The ranges are not a measure of actual participants in the scheme currently.
• Forecasts, by their nature, have an element of uncertainty. The ranges reflect the uncertainty about the participants that will access the scheme, what their needs are, where they will live and what they will choose to spend on what types of support and services. This uncertainty is particularly prevalent in less populated areas, or in areas where the NDIS is not yet available or has not been operating for long. This uncertainty may reduce over time, and the ranges may narrow, as the scheme develops and more data becomes available. The forecasts do not take account of unforeseen changes e.g. sudden changes in demand due to government action
• We expect 90% of postcodes to fall within the reported ranges; but not necessarily the midpoint of the range. This means, 10% of the time, the actual number of participants, participant spending and workforce will be different to what is predicted. It is not possible to say with certainty which postcodes will be different to what is predicted.

Low population areas

• For regions with low populations, caution should be taken when interpreting the ranges. Where the number of participants is expected to be low, an incorrect prediction could be the difference between no spend in a postcode and substantial spend in a postcode. In these cases, confidence could be increased by combining neighbouring postcodes.
• For postcodes with very low populations (less than 50) or small postcodes where the forecast is not reliable, forecast ranges are not provided at this stage. Forecasts may be provided in the future, as more data becomes available, subject to meeting data privacy requirements.
• To ensure privacy, forecast ranges will not be provided for postcodes with participant and workforce populations below 10.

Adding postcodes

• The ranges for individual postcodes will not sum to the estimated ranges if multiple postcodes are selected due to greater confidence when estimating uncertain factors for a greater number of people.
• Website users should not sum categories or sum totals across postcodes, as this will produce inaccurate results.

What is and isn’t captured in the forecasts

• There are some disability services currently funded outside the NDIS that are expected to become part of the NDIS over the coming years. This “in-kind” support does not present an opportunity for NDIS providers at this time but may do so in the future. The forecasts assume that some of this in-kind support becomes part of the NDIS. As this in-kind support actually becomes part of the NDIS, the forecast will become more reliable.
• The model has been adjusted to, as far as possible, capture participants choosing to live in Shared Support Accommodation (SSA). As scheme data is updated and more SSA spend is recorded in NDIS data, the market model will improve the accuracy of spending forecasts associated with SSA.

Workforce estimates

• There is currently no actual data on the number of workers in the NDIS to inform a forecast on the workforce that will be needed in the future. To produce this forecast, the model therefore estimates the workforce through analysis of participant spending, using assumptions on the share of NDIS payments paid as labour costs. These assumptions are based on stakeholder interviews.
• Workforce estimates are provided for a number of occupation groups. Some occupations with low forecast numbers and/or more uncertainty about the specific occupation have been grouped together into “other”. Other may include occupations such as Nurses, Social workers, Podiatrists, Human resource managers, Nutrition Professionals, manufacturers, fitness instructors.

Participant spending categories

• Forecasts of participant spending are broken down across 12 categories. These categories group the 38 NDIS registration groups for simplicity and modelling robustness. Details on which registration group maps to which website category can be found in appendix A.

Other

• The forecast utilises the most recent date postcode data provided by Australia Post. For some postcodes that are situated on state borders, the Australian Post dataset does not recognise postcodes directly bordering in adjoining states. In these instances you will need to directly click on the postcode required to see NDIS demand data.