There are different uses of data in an organisation. Data are required for reporting purposes, decision-making and providing access to vital facts to enable work processes across business units. Data are central to an organisation’s capacity in anchoring fiscal and strategic plans on valid, accurate and current facts, and are also a vital element in an organisation’s capacity to meet legal, compliance and risk management requirements. To ensure sound decision-making, data must be treated as an asset within organisations, with sound data governance principles entrenched and employed for data handling from inception to deletion.
This article proposes a Data Governance Maturity Evaluation Model for government departments of the Eastern Cape province, South Africa.
The methodology for this study is Design Science. The Design Science Process Model, was followed in the development, design and demonstration, and evaluation and communication of the data governance framework. A sequential exploratory mixed-method approach was used for data collection and analysis.
A conceptual data governance maturity model was proposed for government departments of the Eastern Cape province, South Africa. The model was tested through an exploratory sequential mixed-method approach of data collection and analysis. Data were collected from four departments.
The results of the survey confirm the applicability of the model in the set context and reinforced the findings from the literature that maturity models can be used to improve or enhance data governance in public enterprises.
Data serve the diverse needs of different stakeholders in an organisation. These needs include the following: reporting and decision-making; ensuring data quality; data access across organisational divisions; the ability to analyse, sort and filter data; the ability to share sensitive and non-sensitive data in a secure environment; and the capacity to meet legal, compliance and risk management requirements (Dismute
Furthermore, organisations have increased in size and enterprise data have become more complex, with multiple data streams on different devices, personal workstations and bring-your-own-device (BYOD) becoming conventional in the workplace (Alles & Piechocki
To resolve the research problem resulting from the above issues, this article proposes the implementation of a DGMEM for the purpose of managing data assets in the context under review: government departments of the Eastern Cape province, South Africa. This context is apt, as it is home to an estimated 7 million people (STATSsa
Data form the basis of information, which is the central, most important factor employed by government in fiscal and developmental planning. Also, national decision-making, on the one hand, and government budgetary projections, on the other, are heavily dependent on the availability of information, which comes from data collected across a broad spectrum of government departments. It is, therefore, of utmost importance that the data on which such information is based are accurate, valid and complete (Naicker & Jairam-Owthar
The next section discusses the theoretical foundation of this article.
The Data Governance Institute (
… system of decision rights and accountabilities for information-related processes, executed according to agreed-upon models which describe who can take what action with what information, when, under what circumstances, using what methods. (
Seiner (
Based on all the aforementioned definitions, this article adopts a definition of data governance as ‘a homogeneous set of processes which assures formal management of data assets in an enterprise’. The fundamental aim of data governance is to ensure that data are trustworthy, managed by the right human resources and follow a standardised process (IBM
There are various policies and guidelines for IT governance, at both national and provincial levels, which are ostensibly in place for the management and control of IT processes, including data governance and management. Some of these are the Public Service Corporate Governance of Information and the Communication Technology Policy Framework (CGICT) (Department Public Service and Administration [DPSA]
Maturity models are discussed in the next section of the article.
A maturity model is a:
… structured collection of elements that describe the characteristics of effective processes at different stages of development; maturity models suggest points of demarcation between stages and methods of transitioning from one stage to another. (Okongwu, Morimoto & Lauras
Maturity models have also been described as consisting of a sequence of maturity levels for a class of objects. ‘Maturity models represent the anticipated, desired or typical evolution path of these objects shaped as discrete stages. Typically, these objects are organisations or processes’ (Becker, Knackstedt & Poppelbu
The Capability Maturity Model (CMM) is considered the foremost maturity model (Crowston & Qin
The benefits of a Maturity Evaluation Model that serves to highlight the strategic and tactical importance of data governance in government departments of the Eastern Cape are as follows:
The departments will have an organised method of evaluating existing data governance frameworks and policies in a measurable, scientific manner. The model will assist in identifying the gaps, critical skills sets and procedures required to accomplish the data management goals of the departments.
Based on the findings of the evaluation, the DGMEM will systematically outline the processes involved in implementing international data governance best practices and measuring the success of the implementation on well-articulated and clearly defined sets of metrics.
One of the ultimate benefits of a DGMEM is the ability to use the results as a catalyst for building a convincing business case for securing executive sponsorship to support data governance investment and resources for the departments. The model will create an awareness of data governance processes and the negative implications of ungoverned data within government departments.
The institutional and contingency theories form the theoretical basis for this study. The institutional theory was originally proposed by Meyer and Rowan (
The methodology for this study is Design Science, which is increasingly being employed in IS research for creating artefacts to solve ‘real-life’ problems (Peffers et al.
In Design Science, the relationship between people, practice and problems is explored from several viewpoints with the aim of creating an artefact for solving life issues (Johannesson & Perjons
Furthermore, it is argued that the relevance of Information Systems research must stem from its applicability to solving real-life problems, failing which IS research would lose its influence in the field of technology, science and engineering (Peffers et al.
The next section discusses the method of data collection and analysis employed in the study.
Data collection and analysis for this study was conducted using mixed methods. This study employed the exploratory sequential design in the development and validation of the DGMEM through the process of empirical data collection and analysis (Creswell
In the first phase of data collection, open-ended questionnaires, structured in line with the contingency and institutional theories and based on the results of the needs analysis conducted to confirm the existence of the problem, were administered to 45 participants comprising directors, senior managers, middle-level managers, IT managers and data capturers in four selected departments. These have been named D1, D2, D3 and D4, to protect their identities and maintain confidentiality of the information collected. These four departments were chosen for data collection on account of the strategic role data play in their fiscal and operational planning. The 45 respondents form a carefully selected sample, familiar with both IT governance principles and the policy frameworks which reference data handling and management in these departments (Bertram & Christiansen
The qualitative questionnaire was analysed using Nvivo 11 software, and the focus group discussions were analysed thematically. The quantitative data were analysed using the Statistical Package for the Social Sciences (SPSS) version 24 software.
Themes of the Data Governance Maturity Evaluation Model aligned with components of the COBIT 5 and International Organization for Standardization/International Electrotechnical Commission 38500 frameworks.
COBIT 5 | ISO/IEC 38500 | Result of needs analysis in the departments | Correlation between current practices and international frameworks | Themes of the DGMEM to address the lack of data governance processes in the departments |
---|---|---|---|---|
Strategic | Strategic | Unstructured | x | - |
Managerial | Functional guidance | x | - | |
Implementation | Objective evaluation of IT governance processes | x | - | |
Clear information ownership | Accountability | Enterprise data management | √ | Stewardship |
Timely and correct information | - | None | x | Data management lifecycle |
Clear enterprise architecture and efficiency | Ensuring stakeholders are confident of IT and all related governance activities | None | x | Processes, stewardship |
Compliance and security | - | Data security management | √ | Data security and privacy |
Enterprise resources and service capabilities | Evaluate current and future use of IT | None | x | Data management |
IT infrastructure | Ensure the use of IT meets business objective | x | Policies, processes and regulatory compliance | |
People and information | - | x | - | |
Alignment with other relevant standards and frameworks | Monitor conformance to policies and performance against plans | None | x | Regulatory compliance |
Plan – objectives identification, architecture, definition of standards and conventions | Evaluating | x | Policies, processes | |
Design – the physical implementation of what was planned | Directing | x | Data management, processes | |
Build or acquire – covers the creation of data records, acquisition of data assets and data recovery from external sources | Monitoring | x | Security and privacy, regulatory compliance | |
Use or operate – the storage, sharing and disposal of data according to agreed conventions | Directing | x | Data management lifecycle | |
Define data systems | Responsibility | x | Policies, processes | |
CIO determines accountabilities for each layer of data | Strategy | x | Stewardship | |
Data definition, classification, security control and data integrity | Evaluation and monitoring | x | Data security and privacy | |
A nomenclature of where, how and duration of data retention, and clear guidelines on how such data are disposed of or deleted | Performance | None | x | Data management lifecycle |
Unique identification of users, their roles and access levels in tandem with their business roles within the organisation | Human behaviour | None | x | Data management |
Compliance guidelines enforced with enterprise’s contractors or consultants handling data on behalf of the organisation outside its network or firewall settings | Conformance | x | Regulatory compliance |
DGMEM, Data Governance Maturity Evaluation Model; COBIT, Controlled Objectives for Information and Related Technology; ISO/IEC, International Organization for Standardization/International Electrotechnical Commission; CIO, Chief Information Officer.
Following a concise review of data governance literature, government policies and the effective use of maturity models, the components of an effective Data Governance Programme for government departments or public enterprises were determined. These are represented in the DGMEM, as shown in
The Data Governance Maturity Evaluation Model.
The primary focus areas are formulated in line with findings from the literature regarding which aspects of data governance are critical in ensuring accurate data in government departments. The secondary areas were included on the model, as it is believed that the trio of regulatory compliance, metadata management and data management would further assist the departments in maintaining accountability with regard to the verification and accessibility of data assets within the realms of government. The different levels of maturity of the model, as well as the corresponding features inherent in each, are displayed at the top of the model, while the three enablers of stewardship, policies and processes are depicted as the necessary conduit to ensure that the primary and secondary areas of data governance are achievable. The arrows at the base of the model signify Peffers’ process application to the construction of the artefact. The conceptual model is thereafter tested by means of a sequential exploratory mixed-method approach of data collection and analysis.
The process is discussed in the next section of the study.
The three components that form the primary focus area of the conceptual model are data quality management, data lifecycle management, data security and privacy. All of these three elements are at the core of an effective Data Governance Programme. The components also find support in both COBIT and ISO/IEC 38500 (see
The five themes tested in this phase of the questionnaire are people, policies and processes, compliance and management of data assets, data quality, metadata management and alignment of current data processes to international frameworks such as COBIT/ISO/IEC 38500. The same themes were also tested in the focus group discussions. The aim of the exercise is to understand the current data processes in the departments vis-à-vis what is represented on the conceptual DGMEM.
Respondents listed seven different policies as being in place for the governance of data assets in the departments. From the list of policies generated by the responses, it is reasonable to state that there are adequate policies to guard information and, by extension, to manage data in government departments. In spite of this, the findings reveal that end-user data has a vague and unstructured process of management. Furthermore, the findings show that the functional processes around data governance are grossly inadequate to ensure the validity, accuracy and correctness of the data. This finding also negates the recommendation by Korhonen et al. (
Findings from the focus group activities do not seem to negate the results of the qualitative study. However, the nature of the discussions indicates underlying challenges and a severe lack of cohesion in data management within the departments, so much so that provincial offices are far ahead of the regional offices in managing departmental data. It also emerged from the discussions that units in the departments operate in silos and that there is no synthesis in the way data are managed among them. Results revealed a confusion in data roles such as data stewards and data capturers. While respondents stated on the questionnaires that they had dedicated data stewards, the focus group revealed that these were in fact data capturers. One of the unintended consequences of the focus group discussions was finding that most of the senior managers who participated acknowledged the need for changes to data processes and made very useful suggestions on how an enterprise-wide culture of awareness could be driven. This is very important as awareness and buy-in are the first steps to remedial action with regard to instituting a strong Data Governance Programme in organisations (Soares
According to Creswell and Plano Clark (
The next section presents the results of the quantitative analysis of the data.
Reliability analysis.
Variables | Valid |
Items used | Cronbach’s α |
---|---|---|---|
Applicability of DGMEM | 50 | 11 | 0.593 |
Capabilities | 50 | 5 | 0.607 |
Alignment of COBIT | 50 | 13 | 0.677 |
Expected results | 50 | 8 | 0.627 |
Missing components | 50 | 5 | 0.632 |
Note: Applicability of DGMEM is the applicability of the Data Governance Maturity Evaluation Model; capabilities are the people, policies and process capabilities; alignment of COBIT is the alignment of COBIT 5/ISO/IEC 38500 to data process on the DGMEM; expected results are the expected results from the implementation of the DGMEM; missing components are the missing components of the DGMEM.
DGMEM, Data Governance Maturity Evaluation Model; COBIT, Controlled Objectives for Information and Related Technology.
, Significantly acceptable reliability.
A descriptive approach was used to describe the demographic variables of the study. The results indicate that most of the respondents (76%) have been in the departments for over 5 years, which implies that they would be very conversant with the data processes within their units. This is a positive for the study in the sense that results obtained from the questionnaire can be assumed to be based on credible information and experience garnered by respondents over the years.
A one-sample
Applicability of the Data Governance Maturity Evaluation Model (
Number | Statement | Mean | SD | Agree |
|
---|---|---|---|---|---|
% | |||||
1 | The DGMEM is a useful tool to evaluate data governance maturity in my department. (APP1) | 4.30 | 0.46 | 50/50 | 100.0 |
2 | I am able to relate to all the components of the primary focus areas of the DGMEM. (APP2) | 4.24 | 0.59 | 48/50 | 96.0 |
3 | I am able to relate to all the components of the secondary focus areas of the DGMEM. (APP3) | 3.52 | 1.02 | 31/50 | 62.0 |
4 | The processes in the DGMEM for measuring data governance maturity will ensure the verifiability, completeness and accuracy of data. (APP4) | 4.14 | 0.57 | 47/50 | 94.0 |
5 | It is possible to evaluate the maturity level of this department based on this model. (APP5) | 4.64 | 0.53 | 49/50 | 98.0 |
6 | The lifecycle of data will be better managed with the components of the process areas of DGMEM. (APP6) | 4.24 | 0.43 | 50/50 | 100.0 |
7 | Issues of data security and privacy have been adequately addressed by the DGMEM. (APP7) | 3.96 | 0.49 | 45/50 | 90.0 |
8 | The model has addressed and incorporated data management processes relevant to my department. (APP8) | 4.30 | 0.51 | 49/50 | 98.0 |
9 | The importance and management of metadata are adequately covered in the DGMEM. (APP9) | 3.90 | 0.79 | 37/50 | 74.0 |
10 | Regulatory compliance and audit requirements will be met if the DGMEM is implemented. (APP10) | 4.26 | 0.57 | 47/50 | 94.0 |
11 | Both the primary and secondary process areas of the DGMEM present a full picture of our data needs. (APP11) | 4.40 | 0.73 | 47/50 | 94.0 |
Note: Statistically significant differences (*,
DGMEM, Data Governance Maturity Evaluation Model; SD, standard deviation.
Findings from this construct indicate a strong level of agreement with the statements probing the applicability of the DGMEM in the departments. The lowest mean score on this variable was 3.52 (1.02), a clear indication that respondents believe very strongly that the DGMEM is relevant and applicable to their data governance processes.
People, policies and process capabilities (
Number | Do you agree with the following statements? | Mean | SD | Agree |
|
---|---|---|---|---|---|
% | |||||
1 | We have all three components of enabling departmental structures for an effective data governance. (CAP1) | 3.16 | 1.04 | 24/50 | 48.0 |
2 | I believe there are capable human resources to activate the processes in the process model for DGMEM. (CAP2) | 3.08 | 1.10 | 25/50 | 50.0 |
3 | There are adequate policies in place to ensure successful implementation of data governance processes. (CAP3) | 4.00 | 0.83 | 44/50 | 88.0 |
4 | I believe my department is able to achieve the maturity levels based on the process document. (CAP4) | 4.32 | 0.59 | 49/50 | 98.0 |
5 | There are dedicated data stewards to ensure successful graduation from one maturity level to another. (CAP5) | 2.38 | 0.95 | 6/50 | 12.0 |
Note: Statistically significant differences (*,
DGMEM, Data Governance Maturity Evaluation Model; SD, standard deviation.
Alignment of COBIT 5/ISO/IEC 38500 to data processes in the Data Governance Maturity Evaluation Model (
Number | Do you agree with the following statements? | Mean | SD | Agree |
|
---|---|---|---|---|---|
% | |||||
1 | The process areas of the DGMEM match COBIT 5 and ISO/IEC 38500 principles of data governance. (ALI1) | 3.76 | 0.72 | 30/50 | 60.0 |
2 | There is no correlation between COBIT 5 and ISO/IEC 38500 and the work processes in the department. (ALI2) | 2.50 | 0.79 | 8/50 | 16.0 |
3 | We have been adequately trained in COBIT 5 and ISO/IEC 38500 IT/data governance processes. (ALI3) | 2.80 | 1.07 | 18/50 | 36.0 |
4 | I am familiar with the process requirements for data governance in these frameworks. (ALI4) | 3.41 | 1.21 | 30/49 | 61.2 |
5 | There is clear information ownership, as stipulated by COBIT 5, in my department. (ALI5) | 3.16 | 0.91 | 23/50 | 46.0 |
6 | There is a clear and usable enterprise architecture for data efficiency as stipulated by COBIT 5/ISO/IEC 38500. (ALI6) | 2.46 | 0.89 | 7/50 | 14.0 |
7 | The department has a clearly articulated plan–objective–architecture for data governance. (ALI7) | 2.32 | 0.82 | 6/50 | 12.0 |
8 | Enterprise resources and service capabilities, as outlined by COBIT 5, are present in my department. (ALI8) | 2.82 | 0.92 | 16/50 | 32.0 |
9 | The storage, sharing and disposal of data according to agreed conventions are in place. (ALI9) | 3.36 | 0.90 | 31/50 | 62.0 |
10 | Data systems are well designed and well documented in my department. (ALI10) | 2.54 | 0.79 | 8/50 | 16.0 |
11 | There is clear definition, classification and security control of data assets in my department. (ALI11) | 2.38 | 0.83 | 7/50 | 14.0 |
12 | There is a unique identification of users with their access levels in my department. (ALI12) | 3.86 | 0.76 | 42/50 | 84.0 |
13 | Compliance guidelines are enforced with consultants and contractors dealing with departmental data. (ALI13) | 2.39 | 1.15 | 12/49 | 24.5 |
Note: Statistically significant differences (*,
DGMEM, Data Governance Maturity Evaluation Model; SD, standard deviation; COBIT, Controlled Objectives for Information and Related Technologies; ISO/IEC, International Organization for Standardization/International Electrotechnical Commission.
Expected results of the implementation of the Data Governance Maturity Evaluation Model (
Number | Do you agree with the following statements? | Mean | SD | Agree |
|
---|---|---|---|---|---|
% | |||||
1 | The DGMEM will assist us in achieving better data quality. (EXP1) | 4.60 | 0.50 | 50/50 | 100.0 |
2 | Issues of access rights and authentication will be better defined through the DGMEM. (EXP2) | 4.38 | 0.53 | 49/50 | 98.0 |
3 | I believe the DGMEM will encourage the department to seek a higher maturity level. (EXP3) | 4.54 | 0.54 | 49/50 | 98.0 |
4 | The DGMEM will have no impact whatsoever on our data processes. (EXP4) | 1.80 | 0.76 | 2/50 | 4.0 |
5 | Issues of data security and privacy will be resolved if the DGMEM is implemented. (EXP5) | 3.92 | 0.44 | 45/50 | 90.0 |
6 | Regulatory compliance will no longer be an audit problem if the DGMEM is implemented. (EXP6) | 4.14 | 0.54 | 46/50 | 92.0 |
7 | Data management issues can be resolved using the DGMEM. (EXP7) | 4.60 | 0.50 | 50/50 | 100.0 |
8 | The DGMEM will help to create the necessary buy-in for data governance in the department. (EXP8) | 4.16 | 0.79 | 45/50 | 90.0 |
Note: Statistically significant differences (*,
DGMEM, Data Governance Maturity Evaluation Model; SD, standard deviation.
Regarding this variable, the researcher’s intention was to inquire whether the components of data governance in the DGMEM comprehensively cover the essential aspects of data governance as found and discussed in extant literature. To avoid a situation whereby respondents are confused as to the actual meaning of the questions in this variable, it was decided that the best path was to indicate a set of negative statements about missing elements in the model. Four of the five elements of this variable returned a mean of less than 2.60, with the lowest (MIS5) scoring a mean of 2.18. This indicates that respondents strongly disagree with the statements on this variable. MIS4, which states, ‘I believe the DGMEM, as it is, has all the data elements for a successful data governance maturity evaluation suitable for my department’, scored a mean of 4.28, which indicates that respondents strongly believed the statement to be true. Findings regarding this variable align with the qualitative findings (
Missing components of the Data Governance Maturity Evaluation Model (
Number | Statement | Mean | SD | Agree |
|
---|---|---|---|---|---|
% | |||||
1 | There are data elements in my department which are not in this model. (MIS1) | 2.50 | 0.76 | 8/50 | 16.0 |
2 | There are data elements in the model which are not relevant to my department. (MIS2) | 2.30 | 0.79 | 8/50 | 16.0 |
3 | The DGMEM needs to be reconstructed to suit my department’s needs. (MIS3) | 2.18 | 0.48 | 2/50 | 4.0 |
4 | I believe the DGMEM, as it is, has all the data elements for a successful data governance maturity evaluation suitable for my department. (MIS4) | 4.28 | 0.50 | 49/50 | 98.0 |
5 | The DGMEM process model is too complicated to work in my department. (MIS5) | 2.18 | 0.69 | 2/50 | 4.0 |
Note: Statistically significant differences (*,
DGMEM, Data Governance Maturity Evaluation Model; SD, standard deviation.
Descriptive statistics of study variables.
Study variable | Min | Max | Mean | SD | |
---|---|---|---|---|---|
Applicability of DGMEM | 50 | 3.55 | 4.91 | 4.1727 | 0.27930 |
Capabilities | 50 | 2.40 | 4.40 | 3.3880 | 0.57344 |
Alignment of COBIT | 49 | 2.08 | 3.77 | 2.9137 | 0.41324 |
Expected results | 50 | 3.25 | 4.50 | 4.0175 | 0.23147 |
Missing components | 50 | 2.20 | 3.40 | 2.6880 | 0.31079 |
Note:
DGMEM, Data Governance Maturity Evaluation Model; SD, standard deviation.
The one-sample test revealed that only missing components (mean = 2.69,
Graphical depiction of all variables and overall summary of the implications of findings from the data.
One-sample
Overall variable | One-sample |
||||
---|---|---|---|---|---|
Mean | SD | Sig. 2-tailed | |||
Applicability of the DGMEM | 4.17 | 0.28 | 49 | 29.690 | 0.000 |
Capabilities | 3.39 | 0.57 | 49 | 4.784 | 0.000 |
Alignment of COBIT | 2.91 | 0.41 | 48 | −1.463 | 0.150 |
Expected results | 4.02 | 0.23 | 49 | 31.083 | 0.000 |
Missing components | 2.69 | 0.31 | 49 | −7.099 | 0.000 |
Note: Applicability of DGMEM is the applicability of the Data Governance Maturity Evaluation Model (DGMEM); capabilities are the people, policies and process capabilities; alignment of COBIT is the alignment of COBIT 5/ISO/IEC 38500 to data process on the DGMEM; expected results are the expected results from the implementation of the DGMEM; missing components are the missing components of the DGMEM.
SD, standard deviation.
, Statistically lower mean (i.e. less than 3);
, statistically higher mean (i.e. more than 3).
The second component in
Ethical clearance certificate (HER061SOLA01) was obtained from the University of Fort Hare’s Ethics committee before the commencement of data collection for this study.
From the findings presented in
The next section discusses data triangulation and how it was employed to strengthen the results of the inquiry and build confidence in the reliability and replicability of the study.
Triangulation is described by Cohen, Manion and Morrison (
The objective of the DGMEM is to propose the implementation of a DGMEM for the purpose of managing data assets in government departments. The model also presents a process template of how to move from a lower level of maturity to a higher level. The DGMEM is prescriptive in nature, which therefore afforded an opportunity to both the researcher and senior managers in the departments to gain first-hand experience of the model’s ‘fit for purpose’. It is believed that the objective has been achieved as the different layers of data collection and analysis confirm the practicality and applicability of the model to government departments of the Eastern Cape province.
This article discusses the criticality of a sound Data Governance Programme for government departments. A conceptual data governance maturity model is proposed for the government departments of the Eastern Cape province, South Africa. The model was tested through an exploratory sequential mixed-method approach of data collection and analysis. Data were collected from four departments. The results of the survey confirm the applicability of the model in the set context and reinforce the findings in literature that maturity models can be used to improve or enhance data governance in public enterprises.
It is recommended that the model should be implemented for all data processes in a government department as a case study. This will enable the researchers to discover gaps that may be improved to make for more effective data governance. Furthermore, an expanded sample test to other government departments in other provinces of South Africa, or other developing economies, would serve to identify how adaptable the model is to other contexts, and lead to a more robust argument regarding its usefulness and efficacy.
The authors would like to thank the HCD-INTERBURSARY funding body for their support in conducting this research.
The authors declare that they have no financial or personal relationships that may have inappropriately influenced them in writing this article.
O.O. conceptualised and wrote the article, with guidance and input from supervisors. The article is an outcome of a PhD thesis in the stated topic. M.H. is the main supervisor of the doctoral study, who supervised and assisted in synthesising the contents of the article. N.W. co-supervised and offered suggestions to further improve the article.
This research was supported with funding from the CSIR HCD-Interbursary award.
Data sharing is not applicable to this article as no new data were created or analysed in this study.
The views and opinions expressed in this article are those of the authors and do not necessarily reflect the official policy or position of any affiliated agency of the authors.