3.0 Other data issues
3.1 Data frequency
Data frequency is discussed here in terms of modifying the duration between the collection of data for specific data items, in order to re-direct resources toward other survey improvements.
Description of the data issue
External users want data sets to be published at regular intervals, with a focus on supporting time series analysis and the ability to compare with other available data. It is important that data series be up-to-date, so that policy work can be based on current information. It might be cost effective to create a 'data gap' by collecting some data items less frequently, but with a larger sample size when it is collected, allowing for increased accuracy. Some users need annual data therefore it may be prudent to collect selected data items annually at national and state levels, (eg vehicle kilometres travelled, fuel use), and to collect data on other items less frequently (eg data breakdown by vehicle type, or freight data by commodity carried). Analysts may then use the annually supplied data to model annual estimates for data that is available less frequently. This approach would require a transparent schedule for availability of the data gaps.
External users want consultative input in determining the frequency with which individual data items are collected/published. TSUG participants recognise that there is high priority in having data that is accurate and reliable - even if cost considerations mean that the frequency of published data must be reduced.
Questions for stakeholders:
What specifically are external users' data frequency requirements per data item by the following aggregations; national, state, sub-state? Please provide responses within the two tables of Appendix E. The ABS is particularly interested in the following stakeholder data requirements:
- Minimum frequency (data availability) required for key estimates.
- Minimum frequency (data availability) required for specific non-key estimates.
- Please include requirements of other data items not covered by the proforma.
Potential options for addressing data frequency issues
The ABS will consider options available to overcome the issues raised, in response to external stakeholder feedback.
3.2 Capturing transport activity by industry
Description of the data issue (scope)
The current SMVU produces estimates at the national and state levels for the entire economy. Users would benefit from being able to identify transport activity (by transport mode) of each specific industry.
TSUG have suggested identifying industry activity via the reporting of ABNs associated with main operating company of commercial vehicles sampled in the SMVU. The hope here would be to identify the industry code associated with the ABN. A number of methodological issues need to be investigated prior to the ABS determining the feasibility of producing statistically valid SMVU data at the industry level (see below under 'Potential options for addressing transport activity by industry').
Potential options for addressing transport activity by industry
The ABS needs to investigate the conceptual issues, feasibility and cost of deriving industry estimates with the use of ABNs. The costs associated with producing industry based estimates on a yearly basis may be substantial, however industry based 'snap shots' of transport activity (eg every 5 years) may be more feasible. Three potential avenues exist for producing an industry dimension to ABS transport data:
- Obtaining an industry dimension via other ABS collections (discussed in more detail below). Undertake a supplementary survey (asking SMVU specific questions) from an existing industry based survey, producing output by industry. This option may be costly in terms of required ABS resources and provider load.
- Self reporting of ABN by providers (methodological issues relating to this option are identified in more detail below).
- Use of ABN from the survey frames obtained by state motor vehicle registration authorities. The intention here is for the ABS and Stakeholders to work with state motor vehicle registration authorities to assess feasibility of including ABN at point of registration. This would enable a degree of industry stratification for commercially registered vehicles, partially addressing some of the methodological issues. The accuracy of the industry results may be questionable due to the inclination of the industry to engage in leasing arrangements. This may require further investigation prior to determining the validity of the option (further methodological issues relating to this option are identified in more detail below).
If industry based data was only required on a 5 year basis, the sample could be designed and/or boosted every 5th year, to produce industry estimates. Note that this option may have implications on the time series of the major SMVU estimates unless the sample is also designed to control for their accuracy. Any increase in sample size would need to be cleared by the Statistical Clearing House (to assess provider burden on businesses).
Questions for stakeholders:
- How are users currently factoring transport activity within non-transport industries into economic models?
- What costs are associated with this data need not being fulfilled?
- Can the ABS address this data need via other avenues?
- Would data on vehicle counts by industry be sufficient to meet this data need or do users require an industry breakdown of other SMVU estimates?
- Would industry based data be required by users every year, or on an irregular (5 year) basis?
Self reporting of ABN by data providers
The Australian Business Register (ABR) could be used for representational purposes, however the sample could not be stratified or controlled. The following methodological difficulties would need to be investigated and overcome:
- Inaccuracies associated with self reporting of ABN, this is particularly the case for large vertically integrated businesses.
- The same sample representational issues as per the discussion below in 'Use of ABN on survey frames'.
Use of ABN on survey frames
The following short discussion is based on the assumption that obtaining ABN from the survey frames of state motor vehicle registration authorities is feasible. The most expedient method of obtaining an industry dimension to ABS transport data is via the Motor Vehicle Census (MVC). It should be possible to obtain counts of vehicle type by industry, however estimates of kilometres travelled and fuel use would not be possible. The advantage with this avenue is that the methodological difficulties associated with estimation could be avoided.
With ABN on the survey frame, the SMVU could possibly be used to produce statistically valid estimates at the industry level. It may be feasible to produce an industry sub-sample (with the sampling unit being the ABN rather than the vehicle) and to control and stratify the sample on an industry basis. The sampling unit would need to remain the vehicle for the usual SMVU estimates. The following methodological difficulties would need to be investigated and overcome:
- Establishing the correct representational factors (correct sample weights) may be complex for large vertically integrated businesses.
- SMVU sample (The current SMVU sampling unit is vehicle rather than ABN) not representing the number and type of vehicles belonging to the individual ABN.
- A number of ABNs are non operating, but are active (holding companies, trusts). This convolutes the representational factors (correct sample weights) between survey sample and survey frame.
- For some large vertically integrated companies, it is not straightforward which particular ABN relates to the operating business. Moreover it is not always clear which industry the ABN is associated with.
3.3 Cross sector use of transport data (environmental, economic)
3.3.1 Environmental
Description of the data issue (scope)
Environment statistics have an important and emerging role in transport data considerations, with the estimation of pollution emissions becoming increasingly important in research and modelling work. Environmental modellers/researchers require data to enable production of estimates on greenhouse gas emissions (GHG) by industry and households. Researchers require data on road congestion and vehicle noise pollution for the purposes of assessing 'states of the local environment'. Researchers also require measures of the economy's 'eco-efficiency', which encompasses the identification of transport related environmental externalities attributable to economic growth and consumption.
TSUG members do not consider that the ABS has a substantial role other than providing core data that might be used by analysts and modellers. Quantitative description of pollution emissions at low levels of geography was judged by TSUG, to be clearly outside the scope of ABS' national statistics activities. TSUG also considered that measurement of transport wastes is an issue of minor importance and beyond the scope of the ABS transport statistics collections.
Questions for stakeholders:
- Do users have data requirements for environmental research/modelling, that are not currently being addressed by the SMVU?
- How important is the requirement for analysing/combining transport data with social data? Does the SMVU meet the specific data needs for such social analysis?
Potential options to address the issue
Options to address this issue could include:
- ABN based industry data, as per options discussed in 3.2. 'Capturing transport activity by industry' above.
- Other options will be identified if data needs are specified by stakeholders as part of the feedback process.
3.3.2 Measuring financial performance of the transport industry
Description of the data issue (scope)
Users would benefit from being able to identify the correspondence between measures of transport activity with measures of associated financial performance. The absence of data that identifies the in-house transport activity (by transport mode) of non-transport industries, means that modellers cannot accurately forecast the derived physical transport task (by mode) associated with industry forecasts. While noting that the primary emphasis of the SMVU should be transport activity, TSUG have also noted that the ability to link financial data with transport activity data would be desirable.
Questions for stakeholders:
- How are users currently overcoming this data gap?
- Would this data need be satisfied by a supplementary survey, capable of producing SMVU data dissagreggated by industry (one of the proposed options to the 'Capturing transport activity by industry' data issue 3.2.1)?
- What costs are associated with this data need not being fulfilled?
- Can the ABS address this data need via other avenues?
- A transport industry survey would deliver this data requirement. Would stakeholders be willing or prepared to provide funding for such a survey?
Potential options for addressing the financial performance measures data issue
Options to address this issue could include:
- Cover this need by undertaking a transport industry survey. This approach would benefit the ABS by providing National Accounts with transport margin data for input/output purposes.
- Undertake a supplementary survey (SMVU specific) from an industry based survey, producing SMVU output by industry.
The following potential options relate to the possibility of addressing this data need via the use of Confidentialised Unit Record Files (CURF). The ABS needs to investigate the conceptual issues, and feasibility of producing industry based CURFs prior to determining if this option can meet user's data needs. The concern is that data availability would be so restricted, resulting from the confidentialising process, the users needs may not be met.
- Assuming that the methodological issues relating to ABNs are investigated and overcome (as discussed in 3.2 Capturing transport activity by industry), linking via ABN, the SMVU record to an industry collection and provide researcher access to an industry CURF. The likelihood of successful matches to an industry collection may be low and this would need to be investigated.
- Providing researcher access to CURF from a proposed supplementary (SMVU specific) industry survey.
3.4 Vehicle classification used by the ABS
Description of the data issue (scope)
Users of transport data, require that the ABS and other Australian government agencies use common classifications. This will allow for greater comparison and integration of transport data sets produced by different Australian government agencies, including administrative records created by the respective state motor registries. Vehicle type is an SMVU output data item and it is also a stratification variable. The SMVU uses an ABS internal classification for differentiating vehicle type. The ABS is not aware of this classification being used by other organisations.
The classification used by the ABS for the SMVU has 27 vehicle and 8 trailer categories. It focuses on differences in vehicle chassis type and axle configuration. There is an AustRoads vehicle classification that has 12 vehicle categories covering passenger vehicles, rigid trucks and articulated trucks. It focuses on the number of axles and is primarily for assessing freight volume. It has been suggested that components of this classification are useful for users. Note that this classification doesn't have a category for motorcycles, buses or coaches, these categories are required for the SMVU.
The specifics of these classification based data needs are not known to the review team at the present time. The review team will be relying on stakeholders to provide detail on data requirement issues which relate specifically to these classification issues. To facilitate this process, the review team would benefit by stakeholders providing detailed responses to the questions posed immediately below. Copies of the ABS and the Austroads vehicle classifications are located in Appendix F.
Questions for stakeholders:
- The current ABS classification enables users to cross classify data items according to Gross Vehicle Mass (GVM) and total axle number. Does this suit users requirements?
- Are aspects of the Austroads vehicle type classification preferable to the ABS vehicle type classification?
- What additional vehicle type categories not included in the ABS classification would be beneficial (ie aspects of the AustRoads classification)?
- What vehicle type categories currently in the ABS classification are not useful?
- Does the SMVU allow data to be dissagreggated in accordance with users needs?
- Are there any other issues associated with vehicle type classification or other standards that are not covered here?
Potential options to address the classification data issue
Options to address this issue could include:
- Keep the existing classification as is.
- The complete adoption of a different classification used by another organisation.
- Meet user needs by incorporating elements of another classification within the existing classification used by the ABS.
3.5 Question changes on SMVU survey forms
Description of the issue
Improving the non-sampling error (and therefore accuracy) of the SMVU estimates may be achieved by improving the quality of the questions on the survey form. The following survey question issues have been raised and need to be investigated to identify those which may be addressed by the review:
- Given the growing number of passenger vehicles that have 4WD capability, do we need to specifically identify Sport Utility Vehicles (off-road 4WD vehicles)?
- The 'main type of journey' questions need to be structured so as to be unambiguous to the provider, yet align with user requirements. This particular issue requires further investigation by the review team. Work here may achieve better alignment with user requirements than is currently occurring.
- Providers appear to be led by the list of examples used to describe the 'Description of vehicle use' rather than use their own words when answering this question.
- Separate identification of 'low loader' trailers compared to other freight trailers.
- More specific wording on the 'trips per week' question.
- Enhanced clarification and definition of the 'average load weight' question (ie average per trip or average per kilometre).
- Contradiction in the explanatory notes of the 'business use' question.
Questions for stakeholders:
- Do external users need/use 4WD categories? Is the important defining feature drive axle or cabin type?
- Are there any other issues relating to the survey questionnaire which may be addressed by the review?
Potential options for addressing data quality issues
The ABS will consider/develop options available to overcome the issues raised following external stakeholder feedback.