9220.0 - Freight Movements, Australia, Summary, Mar 2001 (Reissue)  
ARCHIVED ISSUE Released at 11:30 AM (CANBERRA TIME) 06/09/2002  Reissue
   Page tools: Print Print Page Print all pages in this productPrint All

TECHNICAL NOTE 1


SAMPLING AND NON-SAMPLING ERROR


INTRODUCTION

1 When interpreting the results of a survey it is important to take into account factors that may affect the reliability of estimates. Such factors can be classified as either sampling error or non-sampling error.

2 Estimates for the road sector in this publication are based on information collected from a sample of registered articulated vehicles, rather than a full enumeration, and are therefore subject to sampling error. The road freight estimates may differ from the figures that would have been produced if the information had been obtained for all registered articulated vehicles. Examples of the sampling error for selected estimates from the FMS 2000 are included in the tables below.

3 It should be noted that although the rail, sea and air sectors of this collection are fully enumerated and are not subject to sampling error, they are still subject to non-sampling error. Non-sampling error is discussed in more detail below.


SAMPLING ERROR

4 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 vehicles 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 vehicles had been included, and about 19 chances in 20 that the difference will be less than two standard errors.

5 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.

6 In this publication, only estimates with a RSE of less than 25% are considered sufficiently reliable for most purposes. Estimates with a RSE between 25% and 50% are preceded by a single asterisk (*) and should be used with caution, while those with a RSE of greater than 50% are preceded by two asterisks (**) and are considered too unreliable for general use (see T1.1 and T1.2 for more detail).

7 The FMS 2000 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 level.

8 The RSEs relating to estimates contained in tables 3 and 4 of the publication are shown in T1.1.

T1.1 RSE OF TONNES CARRIED AND TONNE-KILOMETRES TRAVELLED, ROAD (a)(b)
Destination

Origin
NSW
Vic.
Qld
SA
WA
Tas.
NT
ACT
Aust.

TONNES (%)
NSW
4.51
4.62
5.31
12.06
29.22
. .
45.09
12.57
4.01
Vic.
4.85
3.89
7.25
8.14
26.09
. .
. .
21.13
3.40
Qld
5.80
8.38
5.91
18.74
36.36
. .
25.33
37.86
5.55
SA
8.35
9.47
15.67
7.14
17.37
. .
25.19
40.83
5.98
WA
26.18
29.6
30.86
19.94
6.57
. .
23.90
-
6.52
Tas.
. .
. .
. .
. .
. .
8.25
. .
. .
8.25
NT
74.66
99.59
29.19
22.53
27.31
. .
17.92
-
17.16
ACT
14.92
36.09
52.96
84.36
-
. .
-
30.94
18.17
Aust.
4.06
3.38
5.53
6.10
6.52
8.25
16.37
12.18
2.16

TONNE-KILOMETRES (%)
NSW
2.93
4.01
4.99
9.59
29.60
. .
44.71
12.93
2.09
Vic.
4.18
3.59
7.31
6.79
26.13
. .
-
21.39
2.25
Qld
5.07
9.05
3.84
18.87
37.56
. .
27.33
37.61
2.81
SA
9.44
6.38
16.13
8.27
17.06
. .
21.92
39.47
4.29
WA
26.37
30.44
30.61
18.34
4.56
. .
27.50
-
4.39
Tas.
. .
. .
. .
. .
. .
7.09
. .
. .
7.09
NT
74.66
99.59
35.29
24.13
30.89
. .
13.95
-
10.50
ACT
15.69
36.06
52.15
82.38
-
. .
-
28.20
13.52
Aust.
2.09
2.29
2.65
4.38
4.35
7.09
10.73
10.62
1.39
(a) Data are for freight carried by articulated vehicles only, freight movements made by rigid and light commerical vehicles are
excluded from the survey.
(b) These relative standard errors relate to the estimates of road freight in tables 3 and 4 of the publication.


9 As an example, the estimate of 12,468 million tonne-kilometres moved by road within NSW from table 4 of the publication, has a RSE of 2.93% as shown above (i.e. the standard error is 365 million tonne-kilometres). There are about two chances in three that the figure that would have been obtained if all articulated vehicles had been included in the survey, would have been in the range 12,103 million tonne-kilometres to 12,833 million tonne-kilometres, and about 19 chances in 20 that it would have been in the range 11,738 million tonne-kilometres to 13,198 million tonne-kilometres.

10 RSEs for tonnes carried and tonne-kilometres by commodity (road only), in tables 16 and 17 of the publication are shown below. The RSEs of other detailed variables can be made available on request.

T1.2 RSE OF TONNES CARRIED AND TONNE-KILOMETRES TRAVELLED, BY COMMODITY, ROAD(a)(b)
Tonnes
Tonne-kilometres
Commodity
%
%

Food and live animals
Cereal grains
8.54
5.88
Food (for human and animal consumption)
5.04
3.3
Live animals
6.61
7.25
Total
4.02
2.63
Beverages and tobacco
11.53
8.43
Crude materials, inedible, except fuels
Crude materials
10.48
6.08
Metalliferous ores and metal scrap
16.12
12.2
Sand, stone and gravel
6.87
6.55
Cork and wood
6.27
5.51
Total
4.85
3.45
Mineral fuels, lubricants and related materials
Coal
12.35
11.85
Gases, natural and manufactured
17.88
21.2
Petroleum and petroleum products
7.7
7.89
Total
7.96
6.38
Animal and vegetable oils, fats and waxes
26.49
27.79
Chemical and related products n.e.s.
Chemicals
9.68
10.04
Fertilisers, manufactured
20.98
9.03
Total
11.68
6.83
Manufactured goods classified chiefly by material(c)
Cement
11.99
12.01
Iron and steel
11.3
8.2
Other manufactured goods
7.86
4.97
Total
5.76
4.06
Machinery and transport equipment
5.29
6.71
Miscellaneous manufactured articles(d)
9.24
9.69
Commodities and transactions n.e.s.
General freight(e)
5.49
4.63
Other commodities n.e.s.(f)
8.34
10.69
Total
4.67
4.22
Total
2.16
1.39
(a)

(b)
Data are for freight carried by articulated vehicles only, freight movements made by rigid and light commerical vehicles are
excluded from the survey.
These relative standard errors relate to the estimates of road freight in tables 10 and 11 of the publication.
(c)Manufactured goods used in the production of other manufactured goods and/or made mainly of one material e.g. clay
products, glass and glassware.
(d)Manufactured goods for final consumption and/or made of more than one material.
(e)Consignments not classified by commodity.
(f)All other commodities not elsewhere specified (n.e.s.) including empty used containers, personal effects and furniture not
for sale.



NON-SAMPLING ERROR

11 Other inaccuracies, collectively referred to as non-sampling error, can occur in any statistical collection regardless of whether the collection is based on the full enumeration or a sample. For example, the impact of non-response to the statistical collection, errors in reporting by providers, 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 unrepresentative of the target population. Systematic errors are called bias.

12 Every effort is made to reduce non-sampling error to a minimum by careful design and testing of the questionnaires, efficient operating procedures and systems, and appropriate methodology. Results from the road component of the FMS 2000 were examined for non-sampling error. This resulted in adjustments being applied to compensate for under-reporting. See Technical Note 2 for more detail.