Standard output categories
30. Due to the wide variety of uses of data on age it is not possible to have one standard set of output categories to satisfy all applications. Consequently, the ABS has developed several generally relatable sets of standard output categories. These are based on international standard criteria for classifying age: group size (number of years grouped together), group boundaries (ages defining groups - the boundaries of the groups generally begin at multiples of 5 or 10 and end with numbers ending with the digits of '4' or '9') and age ranges (lowest and highest age of a survey population).
31. The standard output categories are aggregations of the input categories (the level at which age data are collected and stored). The ABS recommends 5 or 10 year age groups for most applications with cut off years varying according to the type of survey (total population or sample), the collection topic, and the user needs and objectives.
32. Categories that adopt 5 year, 10 year and 15 year groupings in the one output presentation are not generally included in the ABS standard classifications, although they are included in the United Nation's International Age Classification. In the majority of cases such combinations of categories appear to be unnecessary and have the disadvantage of distorting the data and impairing objective analysis.
33. In some instances it may be mandatory or highly desirable for output categories to vary from the recommended 5 and 10 year groupings. For example, standard 5 and 10 year groupings may not be suitable for statistical data relating specifically to children or the elderly, or applications where legal requirements such as school leaving age or legal age for marriage are a consideration. Also, general collections relating to deaths need to have their first grouping under 1 year. In such cases variations to the recommended standard output categories are appropriate if they adhere to the general guidelines for age output groupings (see below). However, it is important that such variations are kept to a minimum and are used only where there are compelling reasons for doing so.
34. The following guidelines are recommended when developing output classifications for age:
- The recommended standard output classification to consist of 5 year groupings;
- Groupings with a range of more than 5 years to be in multiples of 5 years;
- Groupings with a range of less than 5 years, including single years, to be capable of aggregation to 5 year groupings as subtotals;
- Subtotals of groupings, and groupings with a range of 5 years or more, to start at numbers ending with the digits '0' or '5' and finish in numbers ending with the digits '4' or '9';
- The terminology used to describe the first group in any set of groupings to be a dash between two numbers (e.g. 0-4), not 'under' (e.g. under 5). However, under special conditions labels such as '4 years and under' are acceptable (see guidelines for published tables below);
- The terminology used to describe the last group in any set of groupings to be the first number of the last age range followed by the words 'and over' (e.g. 75 and over), not the first number of the range followed by a '+' (e.g. 75+).
35. It should be noted that the above recommendations are guidelines rather than rules. For instance, it is not necessary to adhere to these guidelines if there are statistical or other reasons, such as legal age requirements, for deviating from the recommendations. However, as many of the guidelines as possible should be adhered to when developing output classifications or presenting age data. It is particularly important to ensure as many categories as possible are multiples of 5 years, or capable of aggregation to such groupings, to facilitate maximum comparability of age data from different sources.
35(a). The use of 'Age' in a published table may require specific guidelines, in addition to the general guidelines provided above. These guidelines are as follows:
- All totals in tables should be labelled 'Total'. There are some cases where totals are not applicable, as in the case of life tables, which show the calculated longevity of populations at particular ages, rather than a presentation of counts or estimates. As life tables do not tabulate counts or estimates of populations they do not have totals. Although the provision of totals in tables using 'Age' is not compulsory, it is strongly advised.
- Data from sources which have a 'Not stated' code for 'Age' may include a 'Not stated' category in the table. Alternatively, the 'Not stated' response may be incorporated into totals, providing that the 'Total' is footnoted to show it includes these responses.
- Where 'Age' is used in the table stub, the stub heading should be labelled 'Age group' followed in brackets by the unit of time used. The following labels are recommended where appropriate:
- 'Age group (years)'
- 'Age group (months)'
- 'Age group (weeks)'
- 'Age group (days)'.
Units of time such as years, months, weeks etc. should not be mixed in table stubs unless required by the type of data collected. For example, data relating to deaths, particularly infant deaths, may require a combination of time units such as minutes, hours, days, weeks etc. In these cases the stub heading should be labelled 'Age' and the categories should reflect the appropriate unit of time.
- For infants under one year it is not acceptable to use a single category labelled '0', because the appropriate units of time for measuring age of infants are either days, weeks or months rather than years. If a single category is used, it should be labelled 'Under 1 year' (see Specific Requirements below).
- Where the range of ages is restricted by the scope of the survey the output age groups should start at the beginning of the age range and finish at the end of the age range. For example, where the scope of the survey is persons 15-64 years the first age group should begin at 15 years and the last age group should end at 64 years. To ensure the first category starts at the beginning of the age range it may be necessary to use age groups of less than five years for some collections. For example, for a collection on dental health with a lower age limit of 2 years the first age grouping should be '2-4 years', followed by standard five year age groups such as '5-9 years', '10-14 years' etc.
- Labels such as '4 years and under' are only acceptable where the data included in the first category do not necessarily span the full range of the standard category (and the scope of the survey population is not restricted by age). For example in the Schools collection, data can be labelled '4 years and under' because this label refers to children who are under 5 years old. The scope of the collection is not restricted by age but children attending school will not be as young as 1 year. In this case the standard label of '0-4 years' would be inappropriate as it would imply that children less that a year old were attending school.
- The use of an open ended age group at the top of the age range is acceptable when presenting 'Age' output and should be used when referring to the total population derived from sample surveys. The source of the data should be taken into consideration when choosing the top age group as there needs to be adequate sample size to support the group (e.g. '75 and over', '85 and over', etc.)
Standard 5 year groupings
General surveys
36. The standard output categories (excluding Labour Force Surveys) for 5 year groupings are:
0 - 4
5 - 9
10 - 14
15 - 19
20 - 24
25 - 29
30 - 34
35 - 39
40 - 44
45 - 49
50 - 54
55 - 59
60 - 64
65 - 69
70 - 74
75 and over
Total
37. The standard 5 year groupings end at '75 and over' because in a sample survey, any further breakdown for ages over 74 years may cause confidentiality problems and high sampling errors. For example, in small sample surveys the sampling error for the age range 75-79 is typically 12% higher than that of the age range 70-74 years.
Labour force surveys
38. The Labour Force Surveys interview population only include persons 15 years and older with 65 the usual age of detachment from the workforce. Therefore, the standard 5 year groupings for these surveys begin at '15' and end at '65 and over':
15 - 19
20 - 24
25 - 29
30 - 34
35 - 39
40 - 44
45 - 49
50 - 54
55 - 59
60 - 64
65 and over
Total
Administrative collections
39. The standard 5 year groupings for administrative collections cover the total population:
0 - 4
5 - 9
10 - 14
15 - 19
20 - 24
25 - 29
30 - 34
35 - 39
40 - 44
45 - 49
50 - 54
55 - 59
60 - 64
65 - 69
70 - 74
75 - 79
80 - 84
85 and over
Total
40. The groupings '65 - 69, 70 - 74, 75 - 79, 80 - 84, 85 and over' are included because all persons, regardless of their labour force status, are included in these collections.
Census
41. Output produced from the Census uses the above standards as appropriate (eg. output data on the labour force uses the 5 year labour force surveys output groupings).
Standard 10 year groupings
42. The standard output classification using 10 year groupings consists of 11 categories commencing with a 5 year grouping of 0-4 years followed by 10 year categories beginning at 5 years. Inclusion of the 5 year group (0-4) as the first category enables the identification of children of compulsory school age (5-14 years), and persons eligible for participation in the labour force (15 years and over). This is the only situation where a combination of 5 and 10 year groupings is recommended. The classification structure is presented below.
0 - 4
5 - 14
15 - 24
25 - 34
35 - 44
45 - 54
55 - 64
65 - 74
75 - 84
85 - 94
95 and over
Total
43. For survey data the start and end point of the output classification used depends on the nature and purpose of the collection and the sample size employed. For instance, for statistical output relating to labour force participation the first 10 year grouping commences at 15 years and the last grouping at 65 years and over.
Specific requirements
44. It is recognised that some collections have specific requirements which make it mandatory or highly desirable for output categories to vary from the recommended 5 and 10 year groupings. For example, tailored output categories are required for statistical data relating specifically to children, perinatal deaths, the elderly, and applications where legal requirements such as school leaving age or legal age for marriage are considerations.
45. Collections relating to deaths (excluding fetal and neonatal) also need to have their first grouping 'Under 1 Year'. In these cases, variations to the recommended standard output categories are acceptable if they adhere to the general rules for age output groupings (see above). However, it is important that such variations are kept to a minimum and are used only where there are compelling reasons for doing so.
Print Page
Print All