TECHNICAL NOTE DATA QUALITY INDICATORS


DATA QUALITY

1 When interpreting the results of a survey it is important to take into account factors that may affect the reliability of estimates. The survey procedures as well as sampling and non-sampling errors should be considered. Examination of the following quality indicators will assist users in determining fitness for purpose of the Road Freight Movements Survey.

SAMPLING ERROR

2 Estimates in this publication are based on information collected from a sample of registered articulated and rigid trucks rather than a full enumeration, and are therefore subject to sampling error. The Road Freight Movements Survey estimates may differ from the figures that would have been produced if the information had been obtained for all registered articulated and rigid trucks. Examples of the sampling error for selected estimates from the Road Freight Movements Survey are included in the tables below.

3 The sampling error associated with any estimate can be calculated from the sample results. One measure of sampling error is given by the standard error, which indicates the extent to which an estimate might have varied by chance because only a sample of articulated and rigid trucks was included. There are about two chances in three that a sample estimate will differ by less than one standard error from the figure that would have been obtained if all articulated and rigid trucks had been included, and about 19 chances in 20 that the difference will be less than two standard errors.

4 Another measure of sampling variability is the relative standard error (RSE), which is obtained by expressing the standard error as a percentage of the estimate to which it refers. The RSE is a useful measure as it provides an immediate indication of the percentage error likely to have occurred due to sampling. In this release, RSEs are presented in a separate data cube.

5 Estimates with a RSE between 25% and 50% are considered to be high and should be used with caution. Estimates with a RSE higher than 50% are considered unreliable for general use.

6 The Road Freight Movements Survey was designed primarily to minimise relative standard errors for estimates of total tonnes carried, total distance travelled and total tonne-kilometres at the state/territory of registration and vehicle type level.

7 The RSEs relating to some of the estimates contained in table 1 of this release are shown in the following table.

									Destination
Origin	NSW	Vic.	Qld.	SA	WA	Tas.	NT	ACT	Aust.
TONNES (%)
NSW	8.00	10.11	10.62	18.45	47.57	..	65.17	19.37	7.47
Vic.	13.54	13.53	20.55	13.94	64.37	..	..	46.38	12.69
Qld.	11.29	23.68	10.60	42.56	40.62	..	39.68	88.21	10.28
SA	19.10	14.94	34.17	7.25	25.51	..	43.65	..	6.74
WA	52.21	87.80	44.62	26.46	7.42	..	51.81	..	7.41
Tas.	..	..	..	..	..	15.65	..	..	15.65
NT	79.03	..	60.27	51.09	45.34	..	22.27	..	22.10
ACT	22.00	68.77	88.21	..	..	..	..	19.44	16.89
Aust.	7.50	12.85	10.25	6.76	7.41	15.65	22.25	14.17	4.04
TONNE-KILOMETRES (%)
NSW	5.93	10.00	9.26	19.25	48.55	..	64.85	16.94	4.27
Vic.	9.30	8.05	20.27	12.44	64.66	..	..	44.14	5.23
Qld.	11.97	22.62	6.17	30.23	40.93	..	43.96	88.21	5.18
SA	19.34	12.64	40.99	7.05	26.73	..	51.39	..	7.32
WA	57.21	87.55	44.79	25.62	5.46	..	61.37	..	5.41
Tas.	..	..	..	..	..	11.97	..	..	11.97
NT	80.03	..	54.93	60.63	53.81	..	25.90	..	30.36
ACT	23.57	68.94	88.21	..	..	..	..	22.81	14.01
Aust.	4.32	5.39	4.93	7.30	5.36	11.97	29.89	13.41	2.94

(a) Data are for freight carried by articulated and rigid trucks only, freight movements made by light commercial vehicles are
excluded from the survey.
(b) These relative standard errors relate to the estimates of Road Freight Movements, Australia in table 1 of the data cube.

8 As an example of the use of an RSE, the estimate of 30,040 million tonne-kilometres moved within NSW from table 1 of the data cube, has a RSE of 5.93% as shown above (i.e. the standard error is 1,781 million tonne-kilometres). Standard errors can be used to construct confidence intervals around the estimates. There are about two chances in three that the figure that would have been obtained if all articulated and rigid trucks had been included in the survey, would have been in the range 28,259 million tonne-kilometres to 31,821 million tonne-kilometres, and about 19 chances in 20 that it would have been in the range 26,478 million tonne-kilometres to 33,602 million tonne-kilometres (a range of two standard errors above and below the survey estimate).

9 RSEs for tonnes carried and tonne-kilometres by commodity, in tables 6 and 7 of the Road Freight Movements, Australia data cube are shown below. The RSEs of other detailed variables are available within the Road Freight Movements, Australia RSE data cube.

				Tonnes	Tonne-kilometres
Commodity				Australia %	Australia %
COMMODITY RSE (%)
Food and Live Animals
Cereal grains				13.99	10.10
Food (a)				16.06	7.63
Live animals				20.43	14.42
Total				10.03	5.42
Beverages and tobacco				19.80	17.56
Crude materials, inedible, except fuels
Crude materials				35.27	15.26
Metalliferous ores and metal scrap				18.78	15.89
Sand, stone and gravel				11.49	9.34
Cork and wood				15.64	13.47
Total				8.25	6.45
Mineral fuels, lubricants and related materials
Coal				48.53	32.68
Gases (b)				37.03	29.22
Petroleum and petroleum products				17.51	15.55
Total				19.35	12.94
Animal and vegetable oils, fats and waxes				32.98	41.96
Chemicals and related products n.e.s
Chemicals				33.66	18.81
Fertilisers (c)				22.84	15.21
Total				21.68	12.23
Tools of trade				11.98	10.56
Manufactured goods classified chiefly by materials
Cement and concrete				12.07	11.49
Iron and steel				12.26	13.38
Other manufactured articles				11.60	9.97
Total				7.83	6.28
Machinery and transport equipment				11.58	9.00
Miscellaneous manufactured articles (d)				20.39	17.97
Commodities and transactions n.e.s (e)
General freight (f)				7.47	7.12
Other commodity (g)				12.42	10.48
Total				6.38	5.87
Empty				--	--
Total				4.04	2.94

(a) Includes food for animal or human consumption
(b) Natural and manufactured gases
(c) Manufactured
(d) Manufactured goods used in the production of other manufactured goods and/or made of more than one material.
(e) All other commodities not elsewhere specified (n.e.s.) including empty used containers, personal effects and furniture not for sale.
(f) Consignments not classified by commodity.
(g) Not elsewhere specified, including sealed containers and loads.

NON-SAMPLING ERROR

10 Non-sampling error covers the range of errors that are not caused by sampling and can occur in any statistical collection whether it is based on full enumeration or a sample. For example, non-sampling error can occur because of non-response to the statistical collection, errors or omissions in reporting, definition or classification difficulties, errors in transcribing and processing data and under-coverage of the frame from which the sample was selected. If these errors are systematic (not random) then the survey results will be distorted in one direction and therefore will be unrepresentative of the target population. Systematic errors result in bias.

Response and non-response

11 A potentially important factor relating to non-sampling error is the response rate achieved. For the Road Freight Movement survey, the response rate was 77% for vehicles included in the sample.
12 Non-response to the survey predominately occurred because the ABS was unable to trace the selected vehicle or the form was not able to be completed despite all reasonable efforts being taken. Where appropriate, mail reminders and telephone follow-up were used to attempt to contact initially non-responding vehicle owners.

13 For the Road Freight Movements Survey it was assumed that the characteristics of non-responding vehicles were the same as for like responding vehicles. Non-response has the potential to cause non-response bias, which occurs if the usage patterns of the non-responding vehicles differ from those of the responding vehicles.

14 Usable responses were received from 79% of all of the selections for 2014, comprised of 74% from registered vehicles and 5% from unregistered vehicles, out of scope and duplicates. After removing those vehicles that had been found to be deregistered or out of scope, the response rate for the 2014 FMS was 77%.

15 Response rates for each state and territory, and for each vehicle type, are shown in the following tables:

RESPONSE RATES, State/Territory
	Response rate
	%
New South Wales	77
Victoria	74
Queensland	78
South Australia	84
Western Australia	81
Tasmania	78
Northern Territory	70
Australian Capital Territory	72
Australia	77

RESPONSE RATES, Type of vehicle
		Response rate
		%
Type of vehicle
	Rigid trucks	77
	Articulated trucks	78
	Total	77

Frame quality 16 The scope of the survey comprised all articulated and rigid trucks that were registered with a motor vehicle authority for road use at some stage during the 12 months ended 31 October 2014. A population or survey frame was identified using information obtained from the state and territory motor vehicle registration authorities, as part of the ABS Motor Vehicle Census (MVC) (cat. no. 9309.0) for 31 January 2013. From this frame a stratified sample of 16,000 vehicles were selected for reporting on vehicle use. 17 Vehicle classification anomalies arose when respondents indicated an alteration has been made to the vehicle body, resulting in a different vehicle type to that recorded on the frame. These changes happened during the time-lag between finalising the frame and collection of data (between 9 and 21 months). Vehicle classification anomalies can also result from data supplied by state and territory vehicle registration authorities. Imputation 18 Partial imputation is the process whereby a value is generated for missing data items, based on the responses for similar vehicles which were operating for the reference period. As for the Survey of Motor Vehicle Use (SMVU), for which this survey shares its frame, the need for imputation of unanswered items on the returned questionnaires was quite high. 19 Full imputation was used for entire questionnaire non-response. The tables below show the percentage contribution to the estimates from both partial and full imputation. CONTRIBUTION TO ESTIMATES FROM IMPUTATION(a), State/territory of registration Percentage of total tonne-kilometres travelled % New South Wales 32 Victoria 38 Queensland 32 South Australia 27 Western Australia 36 Tasmania 31 Northern Territory 44 Australian Capital Territory 53 Australia 34 (a) Includes both partial and full imputation CONTRIBUTION TO ESTIMATES FROM IMPUTATION(a), Type of vehicle Percentage of total tonne-kilometres travelled % Rigid trucks 38 Articulated trucks 33 Total 34 (a) Includes both partial and full imputation SURVEY PROCEDURES 20 The survey was comprised of three independent samples, with a different one used for each four month period in the overall 12 month survey period. Within each four month period, the sample was divided to cover all 52 weeks of the reference year. Estimates from each of these samples were aggregated to produce annual estimates. Coherent Estimates 21 The results published as Road Freight Movements, Australia (cat. no. 9223.0) in this release are from the Road Freight Movement Survey, which was run as an additional road freight component of the 2014 Survey of Motor Vehicle Use (SMVU). The 2014 SMVU and the road freight component were designed together to provide coherent estimates at the state of registration by vehicle type level for total distance travelled, tonne-kilometres and total tonnes. Adjustments 22 The Road Freight Movements Survey aims to measure the use of all articulated and rigid freight vehicles registered during the reference year. Because selections were taken from vehicles registered some time before the beginning of each collection period, adjustments were made to account for the change in size of the registered motor vehicle fleet since the population frame was created. The frame was created on 31 January 2013 and adjustments involved two categories: re-registrations / de-registrations - typically older vehicles that were either leaving or returning to the registered vehicle fleet after a period of de-registration, and new motor vehicles - vehicles which had not been previously registered. 23 These activities occur continuously and the adjustments were made to account for the registrations that are estimated to have been added to or removed from the registered vehicle fleet between the population frame date and the end of the reference period. The adjustment process also accounted for de-registrations. This means it is possible for the re-registration factor to be negative. CONTRIBUTION OF ADJUSTMENTS FOR RE-REGISTRATIONS, Australia - 2014(a) PERCENTAGE OF TOTAL KILOMETRES TRAVELLED Road Freight Movements, Australia % Type of Vehicle Rigid trucks - Articulated trucks -1 Total - - nil or rounded to zero (including null cells) (a) Data for 12 months ended 31 October. CONTRIBUTION OF NEW VEHICLES REGISTERED AFTER FRAME CREATION - 2014(a) PERCENTAGE OF TOTAL KILOMETRES TRAVELLED Road Freight Movements, Australia % Type of vehicle Rigid trucks 6 Articulated trucks 16 Total 10 (a) Data for 12 months ended 31 October. Pre-advice methodology 24 The quality of survey responses was improved by employing the Survey of Motor Vehicle Use pre-advice methodology. This involved vehicle owners receiving early advice about their inclusion in the survey and encouragement of a higher degree of record keeping. In addition, the reporting of odometer readings taken at the start of the survey period aided in providing reliable estimates of total distance travelled. Nil use 25 Some providers reported nil use for the reference period in which they were selected. Nil use vehicles are registered vehicles that report no travel during that specific reference period. Nil use vehicles were included in the survey as their reported nil use is representative of other vehicles in the population. Vehicles may have nil use due to factors such as seasonal usage, mechanical faults or economic conditions. Where a provider gave a nil use response, a follow-up phone call was used to check the veracity of the response. 26 The survey methodology used for the Road Freight Movements Survey involved a one week collection period. The data collected was therefore more variable than usage data collected for a longer period such as the Survey of Motor Vehicle Use (SMVU) which uses a four month collection period. It was expected that the rate of nil use (and high use) for the Road Freight Movements Survey would be above the rate of nil use (and high use) for the SMVU. 27 The rate of nil use for Road Freight Movements was 16%, compared with 8% for the articulated and rigid trucks segment of the 2014 SMVU.