Inclusion @ Work – Methodology

The findings reported on in this website are based on the nationally representative on-line 2021-2022 Inclusion@Work survey of 3000 Australian workers (except where otherwise indicated). The methodology is described in full detail in the Inclusion@Work Index Full Report, available on DCA website.

Our Approach

  • The methodology was developed and implemented on the basis of ongoing consultation with and feedback from the project’s Expert Panel.
  • It consisted of the following key steps:
  • In-depth review of industry and academic research to investigate how to define and measure workplace inclusion.
  • Development of a draft survey of 70 questions measuring team inclusion, team and individual outcomes, and diversity-related demographics.
  • Pilot of the draft survey in a large organisation and subsequent statistical analysis of the pilot sample dataset to select final 50 survey questions that would generate academically rigorous, practical findings on key inclusion themes (i.e. Respected, Connected, Contributing, Progressing) and key demographics (both social and professional). 
  • Administration of the 2017 survey by Polity Research to a nationally representative sample of 3,000 Australian workers (through a research-only survey panel).
  • Review of the 2017 survey to refine for its second iteration, with consultation involving the expert panel and DCA members that had participated in the Inclusion@YourWork Member Index in 2017. Consultations focused on testing how helpful and appropriate each of the 2017–2018 survey areas were for measuring inclusion, as well as which key demographic areas the survey should include.
  • Review of the 2019 survey in 2021 to refine for the third iteration. To help track results against the prior 2 indexes, changes to the 2021–2022 survey were minimal, involving the removal of 4 survey questions where prior results had demonstrated limitations in the ability to capture meaningful data.
  • Administration of the 2021 survey by Polity Research in May to a nationally representative sample of 3,000 Australian workers (through a research-only survey panel).
  • Weighting of the survey data to ensure the achieved respondent profile aligned with Australian Bureau of Statistics demographic indicators. These include Aboriginality, age, cultural background, disability status, gender, degree qualification, sexual orientation and gender identity, and location (state and urban/regional splits).
  • Analyses of the weighted sample in line with those conducted on the overall sample (e.g., percentage of Australian workers in inclusive, somewhat inclusive and non-inclusive teams) to ensure these findings were representative of the Australian workforce.
  • Crosstab analyses of the unweighted sample comparing the inclusion and exclusion experiences of different demographic groups. These ensured the survey could benefit from the oversampling of key demographic groups including Aboriginal and/or Torres Strait Islander workers, while avoiding the loss of statistical power that can accompany weighting. The crosstab outputs did not appear to differ substantially from their weighted counterparts. 
  • Use of SPSS software to run a series of crosstabs (contingency tables) to identify possible differences in responses. For each contingency table, a chi-squared statistic was calculated to determine whether there was an association between the 2 variables at the 95% confidence level. For contingency tables considered statistically significant, adjusted standardised residuals greater (in absolute size) than +/-1.96 were examined to determine what was driving the association. 

Expert Panel

We thank and acknowledge the project’s Expert Panel. The project has benefited immensely from Panellists generously sharing their expertise and insights.

  • Jonny Ayres, Head of Data Privacy, Novartis
  • Dr Hugh Bainbridge, Senior Lecturer, School of Management, UNSW
  • Catherina Behan, Diversity and Inclusion Manager, People Experience, Suncorp
  • Janin Bredehoeft, Research and Analytics Executive Manager, Workplace Gender Equality Agency
  • Cathy Brown, Policy and Research Manager, DCA
  • Elise Brown, People and Culture Communication Lead, Novartis
  • Bernadette Chehine, Executive Manager, Human Resources, Horticulture Innovation Australia (HIA)
  • Sarah Coombs, Strategy Lead & Company Secretary, Choice
  • Dr Olivia Evans, School of Psychology, University of Newcastle
  • Brigid Furlong, People Experience Advisor – Diversity & Inclusion, Suncorp
  • Associate Professor Dimitria Groutsis, Business School, University of Sydney
  • Dr Beni Halvorsen, Lecturer, School of Management, RMIT
  • Kim Johnson, Head of Organisation Development – Pacific, AON
  • Professor Keith McVilly, School of Social and Political Sciences, Melbourne University
  • Dr Darryl Nelson, Managing Director, Polity Research & Consulting
  • Associate Professor Mark Rubin, School of Psychology, University of Newcastle
  • Emma Schwebel, Human Resources Business Partner, Genworth
  • Dr Graeme Russell, Expert Consultant
  • Professor Phil Taylor, Australian Retirement Research Institute, Federation Business School, Federal University Australia-Monash University
  • Dr Raymond Trau, Senior Lecturer, Department of Management, Macquarie University
  • Lauren Uhlmann, People Experience Advisor, Suncorp
  • Professor Nareen Young, Indigenous Policy (Indigenous Workforce Diversity), Jumbunna Institute, UTS

[1] Workers in Inclusive Organisations scored their team on average at least 4 or above out of 5 on survey questions asking how included they felt in their organisation (where 5 = very included, 3 = neither included or not included, and 1 = not included at all). Workers in Non-Inclusive Organisations scored their team on average less than 3 out of 5.

[1] Where we have indicated that inclusive organisations were “X times more likely” than non-inclusive organisations to be effective, innovative, work hard etc. for ease of reading we have rounded up or down the original number to be a whole number (i.e., 3.5 has been rounded down to 3 times, 4.7 has been rounded up to 5, and 2.9 and 2.8 times have been rounded up to 3 times).