Home > Your ideas > Submissions > UNSW Canberra, Public Service Research Group

UNSW Canberra, Public Service Research Group


No answer

PDF icon Download (246.13 KB)
Automatic Transcription: 
Submission to the Independent Review of the APS: Priorities for Change
(dated 19 March 2019)
2 May 2019

Professor Deborah Blackman, Professor Helen Dickinson, Dr Fiona Buick, Dr Linda Dewey and
Public Service Research Group, School of Business, UNSW Canberra

About us

The Public Service Research Group at UNSW Canberra has a strong track record of research into

public services in Australia and overseas, covering various aspects of public sector management,
delivery of public services and the implementation of public policy1.

Professor Deborah Blackman’s2 research interests include public sector policy implementation,
systems level change, employee performance management, organisational learning, soft knowledge

management, organisational effectiveness, psychological contract, and governance, and she was the

lead researcher on a joint collaborative project with the Australian Public Service Commission on

Strengthening the Performance Framework.3 Professor Helen Dickinson4 has published widely on

governance, leadership, commissioning and priority setting and decision-making in public services, is

co-editor of the Journal of Health, Organization and Management and the Australian Journal of Public

Administration and a Victorian Fellow of the Institute of Public Administration Australia, and has worked

with all levels of government, community organisations and private organisations in Australia, UK, New

Zealand and Europe on research and consultancy programs. Dr Fiona Buick’s5 research focus is on

the value that human resource management can provide to public sector organisations, particularly the

Australian Public Service. She undertakes research into organisational culture, performance

management, and the dynamics involved in structural change, highlighting tensions between informal

and formal institutional practices. Dr Linda Dewey’s research is on the behaviour of the individual in the

workplace. She has worked at senior levels of government and private industry and has worked with all

levels of government on consultancy programs.






Executive Summary

We welcome the opportunity to contribute to this review examining the capability, culture and

operating model of the APS. Under six headings we present our response for implementing

sustainable change in the APS:

• Understanding “culture”
• Culture as an outcome of change
• Shared purpose
• Behaviours and performance evaluation
• Designing a sustainable change program
• The APS as a system

We consider cultural change as an outcome of a change program rather than a driver for change, and

that a clear focus on people that outlines what is changing and why is essential to embedding

sustainable change. A shared purpose and view of the change contextualised to the desired

behaviours at a local level will enable the identification of success prior to the commencement of the

change program. Identification of success enables evaluation of achievement of desired outcomes

rather than ticking off activities. Considering the APS as a system, with programs of work rather than

Departments or Branches as sub-systems enables the identification of existing practices that can be

standardised rather than innovated or deleted to understand what inhibits the embedding of desired


With this approach, we recommend the following approach for consideration by the implementation


  1. Consider the cultural change as an outcome of the change program rather than as a driver
    of the change program
  2. Focus on identifying, initiating and supporting the desired behaviours; through changing
    behaviours, cultural change is likely to evolve over time
  3. Clearly focus on the people involved and be able to describe what is happening and why it is
  4. Develop a shared understanding of the purpose and importance of the public service and
    the value it delivers, and contextualise this as desired behaviours at the local level
  5. Identify and clearly describe what success will look like prior to commencing the change
    program to enable evaluation of the achievement of desired outcomes rather than ticking off
  6. Consider the APS as a system with programs of work, rather than Departments or Branches,
    as sub-systems to enable identification of existing practices that can be standardised rather
    than innovated or deleted through understanding what currently inhibits embedding of
    desired practice.


The Independent Review of the APS: priorities for change (released in March 2019) (Independent

Review) outlines the changes identified to build an Australian Public Service (APS) that will be fit for

the future. The review details the approach adopted by the Review committee, introduces the

subsequent understanding of the challenges, opportunities and aspirations facing an APS

transformation, and includes the priorities for change.

The Review committee is seeking comments on strengthening each proposal, identifying anything

that is missing and recommendations for ensuring lasting change. The recommendations in this

submission are focused on what we believe is necessary to support lasting change.

The Public Service Research Group (PSRG) is broadly supportive of the priorities for change outlined

in the Independent Review and, while we understand the order of the priorities listed in the document

does not necessarily represent a hierarchy of importance, we believe it is important that cultural

change not be held up as the highest priority, nor is it something that predicates the other priorities.
Our experience shows that cultural change, and the accompanying strengthening of governance and

leadership, is an outcome of change rather than a driver of change.

Understanding “culture”

Organisational culture, in everyday language, is considered to be “the way we do things around

here”6. It is the deeply embedded and patterned ways of thinking and behaving evident within groups

and organisations, which evolve over time as a result of shared learning regarding what is most

effective for mission accomplishment7. Because culture evolves over time, and as a result of the

behaviours that have historically led to success, the desire for cultural change must be underpinned

by an understanding of the factors that have led to an undesirable culture in the first place. Therefore,
for this review to move beyond the rhetoric regarding cultural change, and result in real and

sustainable change, we encourage those implementing the Independent Review to approach change

in a fundamentally different way to previous reforms. Despite many waves of reform that have

attempted to change culture within the APS, public service systems as they currently operate in the

APS reinforce recurring issues such as: a culture of risk aversion and silo mentalities, ineffective

people management and individual achievement at the expense of program delivery. Thus, it could be

argued that these reforms have not achieved the desired results. We believe this is because the

assumptions underpinning the current approaches do not facilitate a shift as they have not changed,
despite espoused desires for change over time8.

The Independent Review appears to assume: it is desirable and possible for there to be “one APS”
culture; that such a culture can be clearly identified and achieved; common aspects that are

supportive of the desired performance can be identified; and the existing legislation, structure and

frameworks are supportive of the common culture. It is also assumed that if these aspects are

changed this will lead to improvement, and that a common “one APS” culture is both definable and

achievable. It also appears to be assumed that cultural change is a necessary precondition to

behavioural change; that is, we need to change the culture before more collaborative, open and

transparent behaviours will emerge.

We ask those implementing the Independent Review to consider a different way of approaching this

reform. We suggest that, for there to be real and sustainable cultural change, desired behavioural

changes will need to be identified, initiated and supported first. If these behaviours prove effective and

lead to the attainment of desired outcomes, then it is likely they will result in cultural change; in doing

so, they will become the new norms, with the culture holding them in place in the future.

Deal and Kennedy, 1983, p.501

Schein, 2017

For example, the Management Advisory Committee’s (2004) report on Connecting Government and the Ahead

of the Game 2010 report.

Culture as an outcome of change

With the success rate of change initiatives estimated to be <30 percent9, we suggest approaching this

significant proposed change with careful consideration of the interventions used. Theory suggests the

key aspects of a change strategy include content, people and process10 and highlights the importance

of alignment between these concepts:

• Content includes a clear strategy that links systems, technology and work practices where
technology is key to embedding change; all levels of the business strategies must be aligned
with a strategic and systematic orientation to change11.
• Process is about the implementation of the policies and procedures that are used throughout
the organisation; communication, regular meetings, and the allocation of strong resources
dedicated to the change feature here12.
• Increasing the probability of successful transformation requires attention being given to which
people are involved; addressing their specific needs, and understanding what it is that people
do and how they work to identify what inhibits the embedding of desired practice13.

The key to change success lies in bringing all of these aspects together in a cohesive and collective

way that makes sense to those involved. Previous reforms have tended to concentrate on the

technological and procedural aspects, and the people concept has been primarily concerned with the

espoused desire for cultural change. Such initiatives have not necessarily recognised the complexity

of culture in an entity as large, differentiated and diverse as the APS, where there would be hundreds

of subcultures nested within the broader APS culture. This may mean there has been a lack of

recognition of the plethora of existing, often context bound, subcultures that may operate as driving

forces for change, staunch defenders of the status quo, or quiet (or not so quiet) sabotagers of any

changes or initiatives. Thus, understanding the reality of the current APS cultural landscape, and

gaining consensus from the disparate APS community for the desired culture and how the behaviours

within this culture will be articulated, recognised, and then measured, is a sound starting place for

sustained and sustainable cultural change.

Therefore, we propose the APS reform be implemented using an approach that has a clear focus on

the people involved. We believe the only way to facilitate significant reform is to understand that

humans are central to change and a shift in their behaviour from the current to the desired involves

people identifying with and subsequently altering their own values and perspectives to align with the

overall strategy14. Before this can happen, however, people also need to see why changing their

behaviour is important, particularly for their work group and department. If they can see that these

changed behaviours contribute to successful attainment of desired outcomes, then there is a higher

change they will be maintained. Thus, cultural change becomes an outcome of the change process

rather than a driver of the change process.

Because there are multiple and potentially competing subcultures within the APS that may be

stratified by hierarchy, Department, Division or Branch, it can be assumed that innovations, programs

and interventions will work only in particular circumstances and results are likely to differ for different

groups of participants15. It may be that some of these groups might be readily visible as they are

formed around functions and structure, for example Branch, while others may be latent identifying and

understanding these groups provides a clear framework for implementing change. For this reason, we

argue that creating a strong sense of shared purpose is the first task.

See for example: Al-Haddad and Kotnour, 2015

See for example: Anderson and Ackerman Anderson, 2001 in Al-Haddah and Kotnour, 2015

See for example: Bayerl et al., 2013; Kotnour, 2011; Smith, 2002

See for example: Smith, 2002; Van et al., 2013;
See for example: May and Finch, 2009; Smith, 2002; Van et al., 2013

Moran and Brightman, 2001

See for example: Greenhalgh et al., 2009; Marchal et al., 2012; Pawson and Tilley, 1992

Shared Purpose

We understand the desire for a common purpose in the APS, particularly since it appears there are

two major assumptions underpinning this desire for an APS-wide common purpose: (1) that a

common purpose will enable the APS to become more trustworthy and re-engage with customers;
and (2) that it will make it easier to work across boundaries.

We propose, however, that priority should be placed on establishing a shared understanding of the

importance of public service, but with greater, locally contextual, clarity regarding the purpose and

how this can shape behaviours. The APS is already governed by a set of Values that aim to guide

behaviour in an appropriate way; these values are consistent with the aspirations set out in the

Independent Review for trustworthiness, impartiality, collaboration, accountability, openness, ethical

behaviours, integrity, and a focus on the community16: Therefore, rather than establishing a common

purpose at the level of the APS, we suggest that consideration be devoted to establishing a broad

public service purpose, centred on why the public service exists and the value it delivers to society.
Then this purpose could be made contextually relevant at the departmental level and then localised to

ensure it is meaningful for teams and individuals. This could help drive behavioural change, as

employees can see why exhibiting the desired behaviours is important in a broader sense.

We propose that focus on establishing a common purpose should primarily occur at the local level to

enable collaboration. This is because the capacity of a shared sense of purpose to act as ‘glue’ that

binds people from disparate groups together and enables collaboration, even where there is no

common culture in place, has already been demonstrated17. This would entail ensuring that at the

outset of any policy or program that requires collaboration, attention is devoted to clarifying the

desired outcomes and why the different groups need to work together to achieve it. We argue that

establishing a shared sense of purpose at the level of individual policies and programs will encourage

more collaborative behaviours. There is a possibility that, when operating this way leads to successful

outcomes over time, more collaborative behaviours may become accepted as the norm and,
therefore, valued and embedded in a broader APS culture.

Behaviours and Performance Evaluation

If a desired outcome of the Independent Review is behavioural change, we suggest those

implementing its recommendations identify what success will look like prior to commencing the

change, rather than attempting to evaluate something at the end of the program of work. The

evaluation, and requisite measures, can then focus on desired outcomes in terms of observable


It is acknowledged that measuring change success can take an economic approach (for example:
downsizing, restructuring, redundancies18, efficiency, effectiveness, quality, productivity, innovation,
profitability, budgets) or a capability or human resource approach (for example, leadership, or quality

of work). We agree that defining and setting goals and performance measures is an important

decision, or range of decisions facing organisations during transformational change19. However, we

suggest that evaluation and measurement will be most effective if a blend of approaches are used

that focus on the outcomes in the specific context. This matters because how evaluation and

measurement is framed affects the way the employees will develop their actions. For example, where

there is a desire for avoiding failure, not failing becomes the benchmark that employees’ performance

is measured against; perversely, this contributes to a risk averse culture. In addition, when there is a

desire for high levels of customer satisfaction, this becomes the performance measure; perversely,
this may discourage transparency, integrity and effective governance.

Thus we propose that a clear starting point for sustainable change, would be to identify the aspects of

performance that are currently most valued, affirmed or rewarded within the broader APS community


Buick, 2012

See the following for discussions on types of measurement: Beer and Nohria, 2000; Sink and Tuttle, 1989 in

Al-Haddad and Kotnour (2015)
See for example: Gunasekaran and Kobu, 2007, Itner and Larcker, 1998; Wouters and Sportel, 2005

and the language associated with those aspects. This would help to explain whether they represent

the desired culture or are inadvertently maintaining an inappropriate status quo. This can emerge

through understanding the perceptions and lived experiences of those involved in the change.

Embedded in the discussion of ensuring successful reform is maintaining a focus on outcomes, rather

than ticking off recommendations as they are achieved. A clear vision, supporting strategy and

desired outcomes, including evaluation mechanisms for these outcomes, is integral to the success of

this program. It is important to remain focused on outcomes as concentrating on data can shift the

priority to gathering data, rather than the change program. This results in both a rediscovery of what is

already well established and understood and a strategy of more of something that is already failing to

create the desired systematic and cultural shift. Shifting the focus of measurement to understanding

perceptions and lived experiences of those involved, for example, could provide a very different,
people oriented, way of thinking about and planning for the change.

Designing a sustainable change program

A program approach to transformational change (or reform) is made up of a number of concepts

working together to define the fundamental characteristics of the change required20. Therefore, the

essential first step is articulating or defining the concepts to establish a shared, or a contested, view of
“what the change is”, “who is the supposed target” and “what is the supposed outcome”21. This macro

level of understanding provides a perspective of how the existing structures, cultures, and resources

interact and thus enables a provisional description of what is required of the APS Reform through

describing the principles, characteristics and organisational outcomes associated with the APS

Reform. This macro-level helps identify the way/s people interact and behave because they are
“conditioned” by the current structures and cultures and hence believe their choices are constrained.
For example, are the behaviours associated with procurement decisions within areas of the APS

limited by the FMA Act, the CEIs, procedures developed at a Departmental level or myths that are

active at a local level?

The second step, or meso-level, identifies the new or desired model of practice or the “normalisation”
of what people do. There is often an assumption that a desired model of practice involves the
“innovation” of practices; however, it may also focus on the standardisation of existing practice to

ensure a better practice across a range of situations. For example, the standardisation of

procurement practice or recruitment practice to address transparency. This involves understanding

what it is that people do (rather than what you think they do) and how they work to identify what

inhibits the embedding of desired practice22. For example, if staff in the Department of Human

Services operate within a subculture that is underpinned by the assumption that NewStart recipients

are considered to be “dole bludgers” and “cheats”, the focus of the subculture is catching the cheats

and thus all resources are directed towards this focus and rewarded when they “catch a cheat”.
Changing this model requires significant changes to the prevailing assumptions about human nature,
or at least the intentions of their customers, as well as a significant shift in the use of resources and

the reinforcement (or reward) mechanisms.

Understanding why people behave in a certain way rather than understanding what they are doing

enables the design of interventions designed to shift behaviour and therefore create change. A new

desired practice is made possible when the beliefs, behaviours and acts associated with it have a

shared meaning and use with those who are doing the work, is clearly defined by those doing the

work as being different from the previous practice to establish a new shared meaning for the work,
and this meaning is anchored in the lived experiences of those doing the work. In the case of the

Department of Human Services, a new narrative about NewStart recipients might trigger change far

more than discussing APS values.

Shearn et al., 2017

Pedersen and Rieper, 2008

May and Finch, 2009

The APS as a system

From the above it can be seen that the APS can be conceptualised as a “system” with a series of
“sub-systems”. As such it needs “an integrated approach to drive systematic, constructive change and

minimize the destructive barriers to change, as well as addressing the consequences of making

change”23. We suggest that part of the problem is that the sub systems are currently considered to be

around Departmental lines; hence, for example seeing collaboration as the solution across a silo.

We suggest instead seeing the programs of work (design and delivery) as being the sub-system map.
This will enable a different perspective for the identification of cultures, subcultures and potential

leverage points within the APS; and will enable a more nuanced understanding of what it is that

encourages people act as they actually do24. Each program of work can then be evaluated to

understand the range of resources available within the program of work which includes legislation,
policy, human resources, structural resources, financial resources, external links and so on. From this

analysis it can be determined which resources should remain and be emphasised or built upon, those

that need to change, whether the capability required for this change is already present, and what

mechanisms could facilitate this change.

When using a systems approach, understanding all elements, both tangible (such as technology) and

intangible (such as management capability) is critical in understanding why it is behaving as it is and

what might actually lead to the desired outcomes if it as changed. Understanding the pride and other

cultural elements of the existing APS and harnessing these will be important; as will understanding

the interconnections between these elements, many of which will be operationalised through the flow

of information (both formal and informal) and which links to the earlier recommendation regarding fully

understanding the behaviours linked to why people do what they do – what are the truths and myths

that guide why people do certain things.

Within the system/s there will be stated goals and objectives, however, people will determine

purposes from behaviour they are observing, not the stated goals, and if the behaviour of all players

in the program of work are not aligned to the stated goals, the resulting behaviour may add up to

something no-one wants. There may also be purposes within purposes as systems nest within

systems and the disparate behaviour of players within each of the systems influences the outcome of

each of the sub-systems. For example, within the NDIS system there are procurement, human

resource, information technology and other sub-systems. This all sits within a broader departmental

sub-system which nests within the broader APS system; and each of these sub-systems have their

own behaviour, myths and legends.

Changing the purpose and behaviour of one of the nested sub-systems may result in it being

misaligned with the overall system, thereby altering the performance of the overall system in ways

that were unanticipated and unwanted. Changing the leadership at any level throughout the system

can have the same impact as behaviour drives the change. Changing political leaders results in

changes to the system of the APS; changes to the senior levels within a program of work usually

results in changes to the sub-systems within the program of work. The results are not always


It is also important to understand which feedback loop in which sub-system is dominating and has a

stronger impact on behaviour and program theory enables us to understand what is happening now
(context), to identify some potential changes to this context to understand the mechanisms for change
(the “what if” questions) and to then evaluate what actually happened (outcomes). The APS has been

recruiting staff capable of “resilience” for many years, and this behaviour or ability to “bounce or

spring back into position” may in fact make change difficult as staff are accustomed to managing

difficult change and “bouncing back” after a machinery of government change, a round of productivity

increases or staff or funding decreases. Thus the default setting within the APS seems to be to do

more of the same behaviour. Without fully appreciating what “resilience” is triggering means the APS

Al-Haddah and Kotnour, 2015, p 234.
Lowe and Plimmer, 2019; Meadows, 2008;

may to “bounce back” rather than change and to come up with whole new ways of doing things that

are unpredictable and disruptive or significantly reinforce existing relationships and myths.


In summary, this paper proposes the implementation team consider cultural change within the APS as

an outcome of change, rather than a driver of change. We suggest that focusing on initiating and

supporting behavioural change is likely to contribute to cultural change over time. Through focusing

on the people engaged with the APS as employees, contractors, customers or political actors, the

implementation team can develop a description of what is happening and why.

We propose that the implementation team (and others) develop a shared understanding of the

purpose and importance of the APS and the value it delivers, and contextualise this as desired

behaviours at the local level.

We also propose that the implementation team establish a shared view of what the change is, who it

is supposed to target and its desired outcomes; this will enable definitions of what success will look

like prior to commencing the change program. This description enables the identification of evaluation

criteria and desired outcomes prior to commencing the program so that measurement does not

become a tick-the-box of activities, rather it is focus on changed behaviour and performance.

Finally, we proposed that considering the APS as a system with a series of programs of work as sub-
systems (rather than Departments as sub-systems) will facilitate the identification of existing practices

that can be standardised, rather than innovated or discarded. This will enable the implementation

team to understand what people do and why they do it to uncover what inhibits the embedding of new

or desired practices in the system as a whole.


We recommend the following approach for consideration by the implementation team:

  1. Consider the cultural change as an outcome of the change program rather than as a driver
    of the change program
  2. Focus on identifying, initiating and supporting the desired behaviours; through changing
    behaviours, cultural change is likely to evolve over time
  3. Clearly focus on the people involved and be able to describe what is happening and why it is
  4. Develop a shared understanding of the purpose and importance of the public service and
    the value it delivers, and contextualise this as desired behaviours at the local level
  5. Identify and clearly describe what success will look like prior to commencing the change
    program to enable evaluation of the achievement of desired outcomes rather than ticking off
  6. Consider the APS as a system, with programs of work, rather than Departments or
    Branches, as sub-systems to enable identification of existing practices that can be
    standardised rather than innovated or deleted through understanding what currently inhibits
    embedding of desired practice.


Advisory Group on Reform of Australian Government Administration (2010). Ahead of the Game:
Blueprint For The Reform Of Australian Government Administration. Canberra, Australia:
Commonwealth of Australia.

Al-Haddad, S. and Kotnour, T. (2015) Integrating the organizational change literature: a model for

successful change, Journal of Organizational Change Management, Vol 28 Issue: 2, pp 234-262


Bayerl, P.S. Jacobs, G. Denef, S., van den Berg, R.J., van, Kaptein, N., Birdi, K. and Stojanovski, T.
(2013) The role of macro contenxt for the link between technological and organizational change,
Journal of Organizational Change Management, Vol. 26 No. 5, pp 793-810

Beer, M. and Nohria, N. (2000) Cracking the code of change, Harvard Business Review, Vol. 78 No.
3, pp 133-141.

Buick, F. (2012) Organisational culture as the panacea for joined-up working : the rhetoric and reality

gap. Unpublished Thesis, University of Canberra: http://www.canberra.edu.au/researchrepository/

Deal, T. E. & Kennedy, A. A. (1983). Culture: A new look through old lenses. The Journal of Applied

Behavioral Science. 19(4), 498-505.

Greenhalgh, T., Humphrey, C., Hughes, J., Macfarlane, F., Butler, C., & Pawson, R. (2009). How Do

You Modernize a Health Service? A Realist Evaluation of Whole-Scale Transformation in London.
The Milbank Quarterly, 87(2), 391-416.

Gunasekaran, A., and Kobu, B. (2007) Performance measures and metrics in logistics and supply

chain management: a review of recent literature (1995-2004) for research and applications,
International Journal of Production Research, Vol 45 No. 12, pp 2819-2840.

Ittner, C.D. and Larcker, D.F. (1998) Innovations in performance measurement: trends and research

implications, Journal of Management Accounting Research, Vol. 10 pp 205-238.

Kotnour, T. (2011) An emerging theory of enterprise transformations, Journal of Enterprise

Transformations, Vol 1 No. 1 pp 48-70.

Lowe, T. and Plimmer, D. (2019). Exploring the new world: practical insights for fundsing, commission

and managing complexity. The Tudor Trust: https://collaboratecic.com/exploring-the-new-world-

Management Advisory Committee [MAC] (2004). Connecting government: Whole of government

responses to Australia’s priority challenges. Canberra, Australia: Commonwealth of Australia.

Marchal, B., van Belle, S., van Olmen, J., Hoerée, T. and Kegels, G. (2012). Is realist evaluation

keeping its promise? A review of published empirical studies in the field of health systems research.
Evaluation, 18(2): 192–212.

May, C. and Finch, T. (2009) Implementing, embedding and integrating practices: an outline of

normalization process theory, Sociology, Vol. 43 No. 3, pp 535-554.

Moran, J.W. and Brightman, B.K. (2001) Leading organizational change, The Career Development

International, Vol. 6 No. 2, pp 111-118.

Pawson, R. and Tilley, N. (1997). An introduction to scientific realist evaluation. In E. Chelimsky &
W.R. Shadish (Eds). Evaluation for the 21st century: A handbook (pp. 405-418). Thousand Oaks, CA,
US: Sage Publications.

Pedersen, L.H. and Rieper, O. (2008) Is realist evaluation a realistic approach for complex reforms?
Evaluation, 14, pp 271-293.

Schein, E. 2017. Organizational culture and leadership (5th Ed.). John Wiley & Sons, New Jersey.

Shearn, K., Allmark, P., Piercy, H. and Hirst, J. (2017) Building realist program theory for large

complex ad messy interventions, International Journal of Qualitative Methods, Vol 16, p 1-11.

Smith, M.E. (2002) Implementing organizational change: correlates o success and failure,
Performance Improvement Quarterly, Vol. 15 No. 1, p 67-83

Van, d.H., Demerouti, E. and Bakker, A. (2013) How psychological resources facilitate adaption to

organizational change, European Journal of Work and Organizational Psychology, Vol. 26 No. 6, p 1-

Wouters M. and Sportel, M. (2005) The role of existing measures in developing and implementing

performance measurement systems, International Journal of Operation & Production Management,
Vol. 25 No. 11 pp 1062-1082.