Data analytics and HR

Data analytics is rapidly evolving from a 'nice-to-have' feature to a 'must-have' capability within HR. The potential to uncover actionable insights and make data-driven decisions can offer an unparalleled competitive advantage. In a recent report by WYWM it focuses on the importance of “measure, improve, repeat” and how by collecting data on diversity it can really help to drive change in the areas needed and most likely to strengthen policies in place.[1]?

Nonetheless, this emerging field comes with its own set of ethical considerations. This section navigates the intricate balance between leveraging machine learning for predicting employee behaviour and maintaining ethical integrity.?

Beyond surveys: Utilising machine learning algorithms to predict employee behaviour?

Traditional methods like employee surveys are increasingly seen as reactive, often providing data that is too late to act upon effectively. Machine learning offers proactive solutions, allowing HR to predict and shape behaviours.?

Care should be taken to understand the type of analytics and outcomes that are being carried out to ensure compliance where necessary.??

Automated profiling and decision making under the UK General Data Protection Regulation (UK GDPR) for example, are subject to specific regulations to protect individuals' rights and freedoms. Profiling involves the automated processing of personal data to analyse an individual's behaviour, preferences, or characteristics and using that information to make predictions or decisions about them, while decision making refers to making significant decisions about individuals solely through automated means.?

Benefits:?

Turnover prediction: Algorithms can analyse numerous variables to predict which employees are most likely to leave, enabling preventative measures to be actioned at the appropriate time.?

Performance forecasting: Machine learning can identify patterns that lead to high or low performance, allowing HR to intervene early.?

Implementation guidelines:?

The CIPD[2] outline that “people analytics practice is reliant on good quality, accessible, relevant and reliable data sets that safely and securely store people data and information. CIPD research shows that a barrier to people analytics practice is the over-engineering of complex data systems, under-investment in legacy systems, and over-reliance on overly-sophisticated analytical tools.”?

Below are key steps we recommend all employers can apply when implementing machine learning and analytical tools:?

• Data gathering: The first step involves accumulating relevant data that can be used for training machine learning algorithms.?
• Model selection and training: Once data is collected, the appropriate machine learning model needs to be selected and trained.?
• Continuous learning: As more data is collected, the algorithm needs to be updated to reflect these new insights.?

By incorporating machine learning, HR can shift from a reactive to a proactive stance, elevating its strategic importance within the business.?

Privacy vs Insight: Legal and ethical considerations when employing data analytics in HR?

While the allure of data analytics is compelling, it must be balanced against ethical and legal considerations, particularly concerning privacy and misuse.??

Legal considerations and guidelines?

Legislation: Organisations based in the UK, or who employ individuals who reside in the UK, must comply with the UK GDPR, the Data Protection Act 2018 and other applicable instruments as may become enforced. Employers must ensure that they have an appropriate policy and that privacy information is issued to employees. When considering some of the examples of HR analytics discussed above, it is noteworthy that the UKGDPR prohibits solely automated decisions and profiling when they result in legal or similarly significant effects such as continued employment (unless a legal exemption applies, which is likely to be rare in an employment context). Therefore, it is imperative that your procedures will always be subject to human involvement before any substantial decisions about a person’s employment are actioned.?

International: Where HR analytics entail the processing of personal data internationally, this is likely to fall under the scope of a ‘restricted transfer’. Businesses need to ensure compliance with both the applicable jurisdiction of the countries involved and UK data legislation. This will include ensuring the processing is ‘necessary’ and that the appropriate transfer mechanisms, international data transfer agreements (IDTAs) or binding corporate rules (BCRs), are in place.?

Ethical considerations:?

Data misuse and bias data sets: There is a risk that data could be misused, leading to discriminatory hiring practices or unjust terminations. Furthermore, software may have been pre-trained using data sets subject to systemic bias and prejudices, which could inadvertently lead to discriminatory outcomes if unchecked. Employers should consider establishing a governance board or committee to oversee the use of HR analytics.?

Manjang v Uber is an interesting case which demonstrates the potentially devastating impact of not managing this risk appropriately. Uber deployed face recognition software developed by Microsoft to allow its drivers access to its app whereupon they could access fares. Repeated failed attempts to log-in could lead to automated suspension and removal from the app. This happened to Manjang (and other colleagues of ethnic minority) who subsequently claimed indirect race discrimination. It has been reported by the US National Institute of Standards and Technology that individuals classified in a database as African American or Asian were 10-100 times more likely to be misidentified than those classified as white. The Equality and Human Rights Commission have called out concerns over how this technology is regulated in the UK.?

Employee privacy: As noted above, privacy is a legal consideration and an ethical one too. Advanced analytics could easily turn into over-sharing and even surveillance if not handled carefully. This could create a culture of mistrust as well as legal ramifications if not properly managed. Employees should be informed about what data is being collected and how it will be used (including any new uses) and robust measures must be in place to protect the data from leaks or hacks. This will also help to ensure the legal requirements of transparency and security are adhered to.?

Balancing the advantages of data analytics with ethical imperatives is crucial for maintaining both employee trust and organisational integrity.?

??

This article is an extract from the Whitepaper: Strategic HR Thinking: Aligning people and business strategy: https://www.hrsolutions-uk.com/strategic-hr-thinking-whitepaper/

?[1] WYWM. (2023). The state of workplace Equality, Diversity & Inclusion. [Accessed 16 November 2023]. Available from: https://withyouwithme.com/uk-equality-diversity-inclusion-report/?

[2] CIPD. (n.d.) CIPD viewpoints. People Analytics. Recommendations for employers. Retrieved October 5, 2023, from www.cipd.org/uk/knowledge/factsheets/leadership-factsheet/#differ?

?

?