Brief Summary
This session focuses on preparing HR data for analytics, covering essential steps from data collection to developing HR metrics. It emphasises asking critical questions about data sources, sample characteristics, and potential outliers to ensure data accuracy and reliability. The session also covers data measurement methods, including absolute numbers, ratios, percentiles, and correlations, and stresses the importance of aligning data collection with organisational problems to create effective HR metrics.
- Asking critical questions about data sources
- Understanding data measurement methods
- Aligning data collection with organisational problems
Introduction to HR Data Preparation
The session introduces the importance of preparing HR data for analytics. It highlights the necessity of collecting and transforming data to make informed decisions. The focus is on understanding how to collect data and what preparations to make both before and after data collection to ensure its suitability for analysis.
Key Questions to Ask About Your Data
It's crucial to ask tough questions about your data, regardless of how much you trust the numbers. The first question to ask is about the source of the data, whether it's internal or external. Understanding the source is vital because the reliability of the data depends on the reliability of the source. The second question is whether the sample collected accurately represents the population. If the sample characteristics don't match the population characteristics, predictions about the population will be unreliable. The third question is whether the data distribution includes outliers and how these outliers might affect the results, potentially leading to misleading analysis.
Assumptions and Analytical Approaches
Before analysing data, clarify the underlying assumptions and the reasons for choosing a particular analytical approach, such as descriptive, diagnostic, predictive, or prescriptive statistics. It's important to justify the selected approach and consider available alternatives. Asking these questions about data sources, analysis, and representation enhances the validity and reliability of the data.
Data Measurement Methods
The session discusses how to measure HR-related data, considering whether to use absolute numbers, ratios, or percentiles. Ratios, such as male-to-female ratios, can be presented as percentages. Percentiles, like those used in CAT exam results, indicate a position among a group. Correlation information can reveal patterns, such as the relationship between leader behaviour and employee engagement. The choice of measurement method depends on the data's nature and how it can be presented to provide the most meaningful information.
Data Preparation Aspects
When preparing data, it's essential to check its reliability, accuracy, completeness, variation, and applicability, as well as any missing information. If any of these parameters raise concerns, it's important to address the issues before proceeding with the analysis. Checking the reliability and validity of questionnaires and data sources before data collection is crucial. If a source is questionable, consider alternatives.
Developing HR Metrics
Every Key Performance Indicator (KPI) is a metric, involving numbers that measure something. To develop HR metrics, consider the six main activities of the HR department: recruitment and selection, learning and development, and performance and compensation. Identify the challenges within each process and list the variables. For example, in recruitment, a challenge might be employer branding, which can be measured by tracking how many people have left the organisation versus how many have joined from competitors. Focus on problems the organisation faces and collect data relevant to those issues, rather than wasting time on areas where there are no problems.
Categorising Problems and Developing Metrics
Divide HR-related problems into the six functions and develop metrics accordingly. This categorisation helps create specific recruitment, selection, training, development, performance, and compensation metrics. This approach ensures that data collection and analysis are aligned with the organisation's specific needs and challenges.
Criteria for Good Activity Metrics
When developing metrics, document everything and divide the metrics into the six categories. Each metric should have a narrative definition, explaining any unfamiliar terms. A mathematical formula should be included for every metric, even if it's a simple count. Each metric should explain what is being calculated and why, and should be categorised appropriately. Finally, each metric should indicate the official source of the raw data to ensure transparency and reliability.
