Evidence-based practice: critical interpretation of data for improved decision-making

Dr Sue Shortland explains why it is important to take a critical approach to data interpretation in evidence-based decision-making.

Evidence-based practice: critical interpretation of data.
There is considerable emphasis today on the value of evidence-based practice. In the context of global mobility, this means basing decisions on justifiable and credible sources of information, taking into account relevant theory that predicts likely outcomes.There has been an encouraging growth in research and scholarship in the global mobility arena in recent years and our knowledge and understanding of the various issues that accompany relocation both domestically and internationally have improved considerably as a result. Independent quantitative research, such as surveys, reports and reviews, help mobility professionals to better understand their own organisations through benchmarking with others.With this has come an explosion in the production of surveys and reports on all aspects of the mobility process. Annual benchmarking surveys are commonly produced by mobility service providers. There are also peer-group survey research publications increasingly being produced in response to members’ demands to understand what their colleagues in similar industries and global spheres of operation are doing.While taking mobility decisions based on actual evidence of practice is to be commended, real-world examples represent a historical snapshot, given the incredibly fast pace of change taking place in the world of work. Professional bodies also highlight the inclusion of underpinning theory within evidence-based practice, so that sound decisions can be made regardless of future unknowns. Thus, academic research that advances our understanding of social science theory has a major role to play in practitioners’ understanding of people’s behaviour and decision-making.

The benefits of looking inwards

Evidence-based practice draws upon internal people analytics. Through analysis of their own organisational data, global mobility and HR professionals can learn more about their own workforces, their skills sets, engagement and contribution to business performance. In addition, through the analysis of skills gaps and development needs, strategic investment can be made in human capital. There is the potential for mobility opportunities to be used to the benefit of business outcomes, while employees also benefit from professional and personal development.A further contributor to evidence-based practice is the information gathered from stakeholders who report on their own experiences. Qualitative research is, therefore, valuable as it hears people’s voices. Such research may be based on evidence collected via interviews and focus groups, involving a range of stakeholders. Such data can help to provide evidence on the cultural setting in which an organisation operates, identify capabilities and shortfalls through the posing of questions that require detailed explanations seeking to understand challenges and risks, as well as successes. A nuanced understanding of the stakeholder backdrop provides flesh on the bones of quantitative survey data, enabling identification of future potential options.Global mobility and HR professionals can arm themselves with a wealth of theory, and statistical and qualitative data, but professional judgement is required to make sense of it all and to identify the most appropriate course of action. This requires expertise and a critical approach if effective solutions are to be proposed from the various sources of information available. It is imperative not to take surveys and qualitative research reports at face value without critically scrutinising them. 
Related news

Reviewing evidence critically

Global mobility and HR professionals can take several steps in considering the value of the research data available to them. Two key issues to look for are validity and reliability. Validity refers to whether the findings of the research truly represent the phenomenon being measured. Reliability refers to the potential to repeat the research to gain consistent and stable results.Helpful research studies contain both theory and results, from which it is possible to generalise out to predict outcomes in a particular setting. For example, when examining research data, it is important to look at issues such as sample sizes, response rates and geographies covered. Low sample sizes, response rates and highly specific geographical data can mean generalisation is limited.Widely generalised, all-industry data may also prove to be unsuitable if a more specific focus is required. Data collection techniques, such as reliance on a single source of data, can lead to bias in the findings. Insufficient information on the questions asked means that the findings cannot be contextualised, hampering understanding of the findings and assessing their applicability. Let’s look at some of these issues in more depth and consider examples to help explain these issues.

Samples and sample sizes

It is critically important to check the nature of the sample used in the research study. For example, to consider the input of stakeholders such as expatriates to international assignment policy design, it is necessary to assess whether the sample is drawn from expatriates currently in post, those considering an assignment or repatriates. The currency of their knowledge of international assignment policy and the relevance of their experience to the policy elements being researched must be ascertained and linked to the information required as evidence for policy design decision-making.In examining minorities’ experiences of expatriation to help increase expatriate diversity, it is critical to seek the views of minorities who have undertaken an assignment to find out about facilitators and barriers. For instance, much academic research on female expatriation draws upon Masters level student samples (commenting on the willingness of highly educated women to work abroad), HR managers and academic commentary. Given that academic research has shown that expatriate women have different explanations to those speaking and writing about them, greater insight is likely from the women who have undertaken an international assignment.Small sample sizes are an inherent problem, as large-scale research is costly and time-consuming. The key issue here is to treat the findings with caution and not to generalise out too widely from them. It is also important to consider the sampling design. For instance, in qualitative research ‘snowball’ sampling is frequently used – where one participant refers the researcher on to others. If following the initial contacts, this approach provides a single source of data, this can lead to bias in the findings as the social network used may hold similar views. Random sampling is a stronger form of research design, but it can be difficult to enact due to participant access limitations.

Read more about people analytics data 


Data collection and response rates

If the data collected is based on a single source, there is also a likelihood of bias. Robust research ‘triangulates’ data to improve the validity of the findings. For example, when consulting stakeholders in international assignment policy design, research should consult and obtain data from the range of personnel involved (home and host managers, expatriates, senior strategists, global mobility professionals and so on) to gain a robust understanding of the issues. When examining data presented on policy components produced in survey format by mobility industry suppliers, it is important to assess response rates. While a relatively large number of firms may have participated in the survey overall, response rates to particular questions may not necessarily be high, meaning that the findings are less robust.Much of the data collected in academic and practitioner research represents a snapshot in time – the research design being cross-sectional. This is a function of the time taken to conduct research and participants being less willing to engage in repeated research efforts over time. However, a longitudinal research design can be very helpful in tracking and assessing trends and how outcomes change with time. Many surveys purport to provide data that can demonstrate longitudinal change, but unless there is consistency in sampling (for instance, the same organisations provide input year after year and the questions asked are similar), there is a danger that true comparisons are not possible. This does not preclude discussion of likely trends, but it is important to appreciate that sampling inconsistencies may not give accurate year-on-year comparisons.

Geographical spread

A further point to consider is the potential for geographical bias. Much research – both academic and practitioner – is Western focused. The main regions where research is conducted are in North America and Europe, with specific focus often on the USA and UK.Therefore, the data collected displays geographic and cultural bias. It is notable that when data is included from East Asia, survey findings can change. Japan, for example, tends to adopt the balance sheet, whereas in Europe there is greater usage of host-based or host-plus-based international assignment rewards. When East Asian data is included, expatriate gender diversity tends to be lower. Given the increasing focus on expansion in newly industrialising countries, it is also important to be aware of the paucity of research into expatriation to and from the world’s developing regions.In conclusion, to gain the best outcomes from evidence-based practice, it is essential to review the sources of information with a critical eye. This does not mean dismissing findings, rather it is important and necessary to qualify them appropriately.

For more news and features on data, visit our Technology section.

Subscribe to Relocate Extra, our monthly newsletter, to get all the latest international assignments and global mobility news.Relocate’s new Global Mobility Toolkit provides free information, practical advice and support for HR, global mobility managers and global teams operating overseas.Global Mobility Toolkit download factsheets resource centreAccess hundreds of global services and suppliers in our Online DirectoryClick to get to the Relocate Global Online Directory