Review of 2021 Census Topics

City of Manningham

Extra categories in the Method of Travel to work question and an additional related topic on travel time

Currently Method of Travel to work is collected for the morning of Census day. Our client, the City of Manningham, has suggested that an additional category be added to this classification, to include "Ride Sharing Service" (eg. Uber). This would sit alongside the existing category "Taxi", and reflect the changing nature of these services in the community in the past few years. If Method of Travel to education is included in the Census it could be added to that too.

Additionally, Method of Travel could be enhanced by asking an additional question about travel time. If this could related to "Approximate usual travel time" that would probably be more useful than travel time on a particular day. Currently distance of travel can be calculated using origin and destination information from the Census. But travel time is more important in terms of impact on people's lives, and this can be highly variable depending on location (urban/rural) and mode of travel. Including this topic could help inform national debates which are quite prominent on work/life balance, location of jobs relative to population, and assist in advocacy around better transport options at all levels of government.

Data on travel time are of critical importance at the local level - so a Census is likely to be better than a survey at producing this. Differences in small geographic areas will be the crux of the debate around how long people spend commuting. Small population group info could also be useful, particularly relating to a breakdown by industry or occupation, which would be best assessed by a Census of the whole population. Other demographic characteristics, such as age, sex and income would also be good to correlate with travel time to better understand the choices people make in life.

Adding a category to Method of Travel to Work for "ride sharing service" should be relatively easily collected. Most people are familiar with these services now and should require no additional effort to answer.

For travel time, this will be easily understood in most cases. However the usefulness of the data would probably come from it being asked as a "Usual travel time" rather than a specific occasion (ie. Census day). This could present issues as some people may have travel times which are quite variable, may work in different locations on different days, or not know exactly how long the journey takes them. These issues could be addressed in question wording, and perhaps by including ranges in the question responses rather than a write-in answer. It's unlikely to be perfect but the bulk of the employed population should be able to answer within a reasonable range to provide useful information.

Method of Travel to Work is already collected and is not currently considered particularly intrusive or controversial. Travel Time is likely to be considered less intrusive than many other questions such as income, or even workplace address etc. Given the prominence of debates around work/life balance, the reason for including the question is likely to be obvious to the respondent so it would not seem intrusive or personal. Particularly if it's collected in ranges.

For adding categories to Method of Travel - there may be an issue with increasing the complexity of the output. At the moment every combination of methods is represented up to 3 times in the output categories. So even adding one base level categories will add some 30-50 categories to the result. Consideration should be given to making the output for Method of Travel to Work a multi-response category, like Ancestry - so information about each method can be collected separately, but also cross-classified to provide combination information.

The inclusion of travel time would be adding to the length of the Census as it would need to be an additional question. Unless it could be somehow worked in to the question on work destination. There is also the possibility that it will require a significant amount of instruction in terms of asking respondents to look at an "average" day, concentrate on one place of work etc. - it's not necessary a question which can be answered with minimal thought.

As with many new topics - time series will be of great importance. At a local level transport planners in Local Government and State Government will be able to look at whether their policies are having a direct impact on reducing the journey times of residents, or whether population and jobs growth in different areas is impacting on this and increasing travel times. Cross-tabulation with other Census variables will enable the examination of time series information for small population groups, such as a single occupation category. It seems likely that the topic will remain relevant in the future unless our way of working fundamentally changes - eg. everyone telecommutes - of course this question could monitor that as well.

Alternative data sources for Ride Sharing are a definite possibility. The specific ride sharing companies may have information on the usage of their services, which, if it's not considered commercial in confidence, would provide the number of people using these methods. Of course that would not be integrated with other Census topics or categories of method of travel, which is enhanced by including it on the Census.

For travel time, there may be partial administrative by-product datasets available from state road authorities or public transport authorities. For private travel modes such as car and bicycle, Google clearly have significant data on travel times at particular times of day, and can calculate this through their own service. If this were available outside of Google it would provide a great deal of this useful information for informing the national debate. However it's unlikely to be available and would still lack all the cross-tabulated demographic information which comes with the Census, and will provide the most useful datasets around travel time.

Thanks for the opportunity to submit this on behalf of .id's client base, but particularly the City of Manningham.