A Comparison of Volunteering Rates from the 2006 Census of Population and Housing and the 2006 General Social Survey methodology

Latest release
Reference period
June 2012
Next release Unknown
First release

Appendix A - survey questions

A.1 The 2006 General Social Survey voluntary work questions

The following questions were used in the 2006 GSS.

Q1. The next few questions are about unpaid voluntary work, that is, help willingly given in the form of time, service or skills to a club, organisation or association. Please exclude any voluntary work done overseas.

Show prompt card 36:

  • Organised sporting group / team
  • Youth group (such as guides, scouts, a choir)
  • A charity organisation or cause
  • Student government
  • Religious organisation
  • School or preschool
  • Some other kind of volunteer work

Q2. These are examples of types of organisations or groups that people may do voluntary work for. In the last 12 months, did you do any unpaid voluntary work for any of these organisations?

  • Yes
  • No

Q3. How many organisations have you done unpaid voluntary work for in the last 12 months?

  • 1–14

Q4. The next few questions are about the three organisations for which you worked the most hours in the last 12 months.

What is/are the name(s) of this/these organisation(s)?

Q6. Which of these best describes the type of organisation [organisation] is?

  • Arts / Heritage
  • Business / Professional / Union
  • Welfare / Community
  • Education and Training
  • Parenting, children and youth
  • Emergency services
  • Environment / Animal welfare
  • International Aid / Development
  • Health
  • Law / Justice / Political
  • Religious
  • Sport and physical recreation
  • Other recreation or interest
  • Ethnic and Ethnic-Australian groups
  • Other

Q8. Which groups does the [organisation] aim to assist?

  • Aboriginal / Torres Strait Islander people
  • Aged
  • Children
  • Financially disadvantaged
  • People with disabilities
  • People with chronic / long term illnesses
  • Homeless
  • Ethnic groups
  • Women
  • Drug dependent
  • Youth
  • No specific group / general community
  • Other 

Q9A. Are you currently volunteering for the [organisation]?

  • Yes
  • No

Q9B. How long have you been a volunteer for the [organisation]?

  • Less than one year
  • Between one to five years
  • Between six to ten years
  • More than ten years

Q10. [Is] the reason you assist the [organization]

  • to take part in the Work for the Dole program or Community Work under Mutual Obligation?
  • for work experience, or part of a work trial?
  • because of a student placement?
  • because of emergency work during an industrial dispute?
  • none of the above?

A.2 2006 Census of Population and Housing form questions on voluntary work

The following question was used to collect voluntary work information in the 2006 Census.

Q51. In the last twelve months did the person spend any time doing voluntary work through an organisation or group?

Exclude anything you do as part of your paid employment or to qualify for a Government benefit.
Exclude working in a family business.

  • No, did not do voluntary work
  • Yes, did voluntary work

The 2006 Census Household Guide (page 15) provides the following guidance on answering the question on voluntary work.

Voluntary work

Q51. Only include help willingly given, in the form of time, service or skills, to a club, organisation or association.

Unpaid voluntary work can include:

  • assisting at organised events and with sports organisations
  • helping with organised school events and activities
  • assisting in churches, hospitals, nursing homes and charities
  • other kinds of volunteer work (eg. emergency services, serving on a committee for a club etc.)

If the person is doing unpaid voluntary work through a club, organisation or association in order to qualify for government benefits such as Newstart Allowance, do not include this as voluntary work at Q51.
Do not include any activity that is part of the person’s paid employment, or family business.

Appendix B - survey methodology and proxy reporting

The 1995 Voluntary Work Survey run by the ABS had a randomly selected adult responding on behalf of the whole household. Previous analysis on this survey and method of data collection (ABS, 2000) indicated that the people personally interviewed were much more likely to report volunteering than when volunteering information was collected about the person from someone else in the household. The difference in reporting was about 5:1, with a voluntary work rate of 27% for those who responded on behalf of themselves and 5.5% when their voluntary work was reported on by someone else. This is equivalent to reporting the difference between self-report and ‘Any Responsible Adult’ methodology. For more information see Voluntary Work, Australia, 2000 (ABS, 2000).

This analysis proposed that lower volunteer rates in the Census could be largely due to one person responding on behalf of all household members and under reporting voluntary work due to lack of knowledge on voluntary activities about other members of the household. This effect is known as a ‘proxy reporting’ effect. Proxy reporting is an issue that could have occurred in the Census but not in the GSS (barring exceptional circumstances of proxy reporting) due to the different collection methodologies used.

It is not possible to directly test for proxy reporting in the Census because there is no way of knowing which person on the Census form responded on behalf of others. To analyse proxy reporting, lone person households were analysed from both collections (restricting the analysis to the common population), as they are assumed to have necessarily completed the Census on their own behalf in most circumstances (some proxy reporting may occur when someone outside of the household, such as a family member, may complete the Census on behalf of a lone person) and therefore would have reported their own volunteering behaviour. This means there is unlikely to be a proxy reporting effect for this type of household. If the volunteer rates are still different between the two collections we can assume other effects influencing the difference as well, such as the collection methodology or question content.

In the GSS, lone-person households had a volunteering rate of 26.9% and the Census had a volunteering rate of 20.1%. There is still a difference in the voluntary work rate of around 6.8 percentage points between the two collections, so proxy reporting cannot completely account for the difference between the GSS and the Census.

Appendix C - logistic regression coefficients

C.1 Logistic regression coefficients predicting propensity to volunteer, 2006 GSS
95% confidence interval
CharacteristicsCoefficientStandard errorWald Chi-squareP > ChiSqLowerUpper
Male (base)0.000
Age group
18-29 years-0.3790.0642.530.00-0.49-0.26
30-39 years-0.2270.0518.470.00-0.33-0.12
40-59 years (base)0.000
60 years and over0.0850.061.920.17-0.040.21
Family composition
Family with children aged under 15 years0.7110.05218.470.000.620.81
Family with dependent children aged 15-24 years0.3320.0815.650.000.170.50
Family with no dependent children (base)0.000
Social marital status
Married (base)0.000
Not Married-0.1440.0412.440.00-0.22-0.06
Proficiency in spoken English at home
Speaks English only (base)0.000
Speaks English well or very well-0.4320.0741.710.00-0.56-0.30
Speaks English not well or not at all-1.2930.1850.570.00-1.65-0.94
Highest level of educational attainment
Advanced Diploma/Diploma or above0.8830.05319.190.000.790.98
Year 12 or Certificate III/IV0.4190.0574.410.000.320.51
Year 11 or below (base)0.000
Labour force status
Employed full-time (base)0.000
Employed part-time0.2570.0522.450.000.150.36
Not in the Labour force-0.0610.061.180.28-0.170.05
Part of State
Capital city-0.2320.0432.030.00-0.31-0.15
Rest of State (base)0.000
  1. Married includes registered marriages and de-facto relationships where the couples live together. Not married includes all other living arrangements.
Model diagnostics
Number of observations13,227
Likelihood ratioChi-Square = 963.86
P > ChiSq < 0.0001
Hosmer and Lemeshow Goodness-of-Fit testChi-Square = 10.71
P > ChiSq = 0.22

Appendix D - odds, odds ratios, and predicted probability


The odds of an outcome is the ratio of the expected number of times the event will occur to the expected number of times the event will not occur. Put simply, the odds are the ratio of the probability of an event occurring to the probability of no event. In our case, we can calculate the odds that our base case is a volunteer according to the following ratio:

\(\begin{array}{c} \large {\text { Odds for volunteering }=\frac{\text { Probability of being a volunteer }}{\text { Probability of being a non-volunteer }}}\\ \end{array}\)


Applying a logistic regression to the GSS data, the odds of being a volunteer for the base case is estimated at 0.35.

Odds ratio

This is the ratio of two odds. In our example, the odds ratio compares the odds of volunteering by a female to the odds of volunteering by a male. Imagine a person who has all the same characteristics as the base case except that she is female. To work out the effect of being female on being a volunteer we can calculate an odds ratio using the odds for each of the cases as follows:

\(\large\text{Odds ratio for female}=\frac{\text { Odds that a female is a volunteer }}{\text {Odds that a male is a volunteer }}=1.28\)

If the odds ratio equals 1, then men and women are equally likely to be volunteers. An odds ratio of less than 1 would suggest that, all other things being equal, women are less likely than men to be volunteers. In our case, the odds ratio for women is 1.28 suggesting that women are more likely to be volunteers than men. Put differently, a woman’s odds of volunteering are 28% larger than a man’s.

Predicted probability of volunteerism

Once we obtain the odds ratio, we can convert this odds ratio into predicted probability by simple algebraic transformation. The relationship between the odds ratio and probability is as follows:

\(\large\mathrm{Odds}=exp{\left(\beta_{0}+\beta_{1} x_{1}+\ldots+\beta_{p} x_{p}\right)}=\frac{P[Y=1]}{1-P[Y=1]}\)

Simplifying the notation:


Cross multiplying and rearranging the above:

\(\large\begin{aligned} \text { Odds }-P \times \mathrm{Odds} &=P \\ P+P \times \mathrm{Odds} &=\text { Odds } \\ P(1+\text { Odds }) &=\text { Odds } \\ P &=\frac{\text { Odds }}{1+\text { Odds }} \end{aligned}\)

Continuing from the example above, the odds for a female to be a volunteer is:

\(\large\begin{aligned} {\text { Odds for volunteering }} & \times \text { Odds ratio for female } \\ & \ {= 0.35 \times 1.28=0.45}\end{aligned}\)

This odds then can be translated into the predicted probability by:

\(\large{P}=\frac{\text { Odds }}{1+\mathrm{Odds}}=\frac{0.45} {1+0.45}=0.31, \text { or } 31 \% \)

Appendix E - 2006 Census volunteer rates for all Sydney SLAs

E.1 Volunteer rates, all Sydney Statistical Local Areas (SLAs)(a) – 2006 Census
Volunteer Rate (%)Volunteer Rate (%)
Fairfield (C) - East7.9Canada Bay (A) - Concord16.9
Bankstown (C) - North-East9.0Sydney (C) - South17.1
Fairfield (C) - West9.1Wyong (A) - South and West17.4
Parramatta (C) - South9.3Parramatta (C) - North-West17.5
Liverpool (C) - West10.1Sydney (C) - West17.6
Canterbury (C)10.3Sutherland Shire (A) - East17.8
Auburn (A)10.4Sydney (C) - East17.8
Blacktown (C) - South-West10.5Ryde (C)18.8
Botany Bay (C)10.6Parramatta (C) - North-East19.1
Bankstown (C) - North-West11.2Leichhardt (A)19.2
Liverpool (C) - East11.2Waverly (A)19.3
Rockdale (C)11.3Gosford (C) - West19.5
Holroyd (C)11.9Baulkham Hills (A) - North19.7
Penrith (C) - East11.9Wollondilly (A)19.7
Parramatta (C) - Inner12.1Sutherland Shire (A) - West19.9
Sydney (C) - Inner12.6North Sydney (A)19.9
Blacktown (C) - South-East12.8Hawkesbury (C)20.4
Campbelltown (C) - North13.4Baulkham Hills (A) - Central20.5
Campbelltown (C) - South13.8Warringah (A)20.6
Penrith (C) - West14.4Willoughby (C)21.0
Blacktown (C) - North14.4Gosford (C) - East22.4
Kogarah (A)14.6Baulkham Hills (A) - South22.7
Hurstville (C)14.8Woollahra (A)22.8
Burwood (A)15.2Pittwater (A)23.1
Bankstown (C) - South15.5Manly (A)23.2
Wyong (A) - North-East15.7Lane Cove (A)23.9
Strathfield (A)16.0Hornsby (A) - North24.0
Ashfield (A)16.3Mosman (A)24.7
Marrickville (A)16.4Hornsby (A) - South24.9
Canada Bay (A) - Drummoyne16.5Hunter's Hill (A)25.2
Randwick (C)16.5Blue Mountains (C)27.1
Camden (A)16.7Ku-ring-gai (A)29.2
  1. To keep the analyses consistent, Census estimates presented in this table are for people aged 18 years and over living in private dwellings, in non-very remote areas, with a stated voluntary work status and no missing values.


Show all

Back to top of the page