6227.0.55.002 - Experimental estimates of education and training performance measures based on data pooling, Survey of Education and Work, 2007 to 2010, Sep 2011  
ARCHIVED ISSUE Released at 11:30 AM (CANBERRA TIME) 09/09/2011  First Issue
   Page tools: Print Print Page Print all pages in this productPrint All

NASWD 2: LOW QUALIFICATIONS, 20-64 years

Compared with the other indicators, NASWD 2 (low qualifications, 20-64 years) is based on a very large sample. In the 2010 SEW there were around 35,000 people aged 20-64 years in the sample of whom about 16,000 had qualifications below Certificate III (Datacube Table 10). Furthermore, NASWD 2 changed more rapidly than the other measures examined, particularly the other attainment indicator, NEA 7 (Year 12/certificate II, 20-24 years). NASWD 7 fell by about five percentage points between 2007 and 2010 whereas NEA 7 rose by only two. The continuing decline in the proportion of people with qualifications below Certificate III reflects generational change, whereby older age cohorts with, on average, relatively low qualification levels are being replaced by younger ones with higher qualifications. It may also reflect, in part, access across all ages to continuing education and training opportunities.

Even so, this indicator may benefit most from data pooling to double or even quadruple the sample. On the one hand, combining data over successive years led to mixed benefits, since change over two years could be detected with a similar degree of frequency using either pooled or single year data. On the other hand, data pooling over two years did lead to detection of change at the state/territory level for most jurisdictions and to some detection of change at the national level by 10-year age group or SEIFA quintile.

The discussion of this indicator is similar to that for the other attainment indicator, NEA 7 (Year 12/Certificate II, 20-24 years) except that the population base of 20-64 year olds is much larger and the size and rate of change much greater. Indeed, single-year data for NASWD 2 in 2010 was based on a larger sample (denominator) than that for any of the other indicators even after four-year pooling (Datacube Table 10).


NATIONAL AND STATE/TERRITORY DATA

At the national level, the proportion of 20 to 64 year olds with qualifications below Certificate III fell from 50.1% in 2007 to 45.4% in 2010 (Datacube Table 1a).

Unlike the indicators measuring participation in education/training and employment, such as NEA 9 (engagement, 15-19 years) and NEA 10 (engagement, 18-24 years), it is not strongly affected by the economic cycle. At the national level, RSEs in 2010 ranged from 0.6% for single and two-year pooled data to 0.4% for four-year pooled data (Table G below). At the national level, the accuracy of this indicator was adequate for measuring annual change (falls) in the proportion of people aged 20-64 years with qualifications below Certificate III between 2008 and 2009 and between 2009 and 2010. Accordingly, single-year and pooled data also recorded statistically significant change over the two years 2008 to 2010 (Datacube Table 1c) .

At the state/territory level, RSEs for single-year data in 2010 ranged from 1.4% to 2.1% in the five largest states: NSW, Victoria, Queensland, SA and WA, and from 3.0% to 3.8% in Tasmania, the NT and the ACT. For the smaller states and territories, RSEs were reduced to below 3% under two-year pooling and around 2% under four-year pooling (Table G below). Virtually no statistically significant changes in attainment were observed over the one-year periods 2008 to 2009 or 2009 to 2010 - the exception was a fall in Victoria between 2009 and 2010 (Datacube Tables 3c). Data pooling over two years led to greater detection of change over the two-year period 2008 to 2010 but these changes were also largely detectable using single-year data. Between 2008 and 2010, both single-year and pooled data recorded six out of eight jurisdictions with significant falls in the indicator (even though the set of jurisdictions differed slightly) (Datacube Tables 2c, 3c, 4c, 5c, 6c, 7c, 8c).

TABLE G. DIFFERENCES IN ESTIMATES FOR 2010, change through data pooling (NASWD 2)

Single-year

(2010)
2-year pooled

(2009-10)
4-year pooled

(2007-10)

'000
%
RSE of %
'000
%
RSE of %
'000
%
RSE of %

NSW
1841.8
44.2
1.4
1865.9
44.8
1.1
1923.1
46.2
0.8
Vic.
1445.5
44.0
1.5
1502.6
45.7
1.2
1540.4
46.9
0.9
Qld
1244.1
47.5
1.6
1250.3
47.7
1.6
1289.8
49.2
1.1
SA
468.9
49.2
2.1
477.2
50.1
1.5
490.6
51.5
0.9
WA
634.8
46.5
1.9
646.6
47.3
1.4
666.7
48.8
1.1
Tas.
144.3
50.5
3.0
145.1
50.8
2.4
149.3
52.3
1.6
NT
64.1
47.0
3.5
65.9
48.2
2.6
69.1
50.6
2.1
ACT
74.6
33.9
3.8
77.7
35.3
2.8
80.3
36.5
2.1
Australia
5918.1
45.4
0.6
6031.3
46.3
0.6
6209.3
47.6
0.4



AGE

By age, the proportion of 20-64 year olds without at least a Certificate III was highest for the youngest age group, 20-24 years, many of whom were still studying and had yet to attain a qualification. Excluding this age group, the proportion of the population without a Certificate III or higher qualification generally increased with age reflecting growth in qualifications by successive age cohorts over time. At the national level RSEs for single-year data in 2010 ranged between 1.0% and 1.9% across the 10-year age groups (Datacube Table 1a).

Data pooling did not lead to an appreciable benefit in measuring change in this indicator by age group. At the national level, change by age group was sometimes detectable over one year and largely detectable over two years, using either single-year or pooled data (Datacube Table 1c). At the state/territory level, virtually no age group comparisons were detectable over the one-year periods 2008 to 2009 or 2009 to 2010, with the exception of Victoria where three age-groups recorded statistically significant falls in low attainment between 2009 and 2010 (Datacube Table 3c). A small number of state/territory comparisons were significant over the two-year period, 2008 to 2010, even in the NT and ACT, but without a consistent pattern between results for single-year and pooled data (Datacube Tables 2c, 3c, 4c, 5c, 6c, 7c, 8c). Some of these results may have been due to sampling variability rather than actual change. See the section on significance testing for further explanation of sampling variability and measures of change.


SOCIOECONOMIC STATUS

Since this indicator measures people with low qualifications, it is relatively high in the lower SEIFA quintiles and lower in higher quintiles [1]. In 2010, RSEs for single-year data at the national level ranged from 1.3% to 1.9% (Datacube Table 1a). Again, data pooling did not lead to appreciable benefits in measuring change in this indicator by socioeconomic status over the short-term. No change by SEIFA quintile was detectable between 2008 and 2009 or between 2009 and 2010. Over the two-year period 2008 to 2010 some changes were statistically significant based on either single-year or pooled data (Datacube Table 1c).

When pooled data was examined over 2008 to 2010 at the state/territory level, change was detectable in one or two quintiles for most states/territories, although given the small sample size some of these results may have been due to sampling variability rather than actual change. In contrast, there were few detectable changes over the one or two-year periods using single-year data (Datacube Tables 2c, 3c, 4c, 5c, 6c, 7c, 8c).


FOOTNOTE

[1] SEIFA Index of Relative Socioeconomic Disadvantage. For further information on this and other SEIFA Indexes, see Information Paper: An Introduction to Socio-Economic Indexes for Areas (SEIFA), 2006 (cat. no. 2039.0) (back to text)