![So, from the metrics- way we have now come to applying HR
Analytics,
which comprises :
ļµ A methodology for developing innovative insights
ļµ Linking HR data, survey data, and business or organization data
(financial, client, operationalā¦)
ļµ Applying analytical and machine language techniques ( Factor
and Cluster analysis, correlations, regression, Logistic regression,
Classification trees modeling, using natural language programming,
text mining, word cloudsā¦);
ļµ Quantifying (or identifying) the impact of investments in human
capital on the business or organizationā.
[4 points Definition (the emphasis is on Impact)]](https://image.slidesharecdn.com/hranalytics-250214082719-fd379f65/85/This-ppt-is-about-HR-Analytics-and-its-applications-14-320.jpg)
This ppt is about HR Analytics and its applications
- 1. HR Analytics
- 2. HR Analytics: Transforming HR with Data-Driven Insights ļµ Human Resources (HR) analytics, also known as workforce analytics peoples analytics or talent analytics, is the use of data analysis techniques to understand, improve, and optimize HR processes and decisions. By leveraging data, HR professionals can gain valuable insights into their workforce, enabling them to make more informed decisions and drive strategic initiatives.
- 3. ļµ HR analytics is a powerful tool that can help HR professionals transform HR processes and drive strategic initiatives. By leveraging data-driven insights, HR can make more informed decisions, improve employee engagement and retention, and align HR initiatives with overall business goals. As organizations continue to recognize the importance of their workforce as a strategic asset, HR analytics will play an increasingly critical role in shaping the future of work.
- 4. Steps in HR Analytics: ļµ Data Collection: HR analytics begins with the collection of relevant data from various sources, including HR systems (e.g., HRIS, ATS), employee surveys, performance reviews, and external sources (e.g., market data, social media). ļµ Data Cleaning and Preprocessing: Once the data is collected, it needs to be cleaned and preprocessed to ensure accuracy and consistency. This may involve removing duplicates, handling missing values, and standardizing data formats. ļµ Data Analysis: The next step is to analyze the data using statistical techniques and machine learning algorithms. This can help HR professionals uncover patterns, trends, and relationships in the data. ļµ Data Visualization: Data visualization techniques, such as charts, graphs, and dashboards, can help HR professionals visualize the data and communicate key insights effectively. ļµ Predictive Modeling: HR analytics often involves building predictive models to forecast future trends and outcomes, such as employee turnover, performance, and engagement. ļµ Benchmarking: Benchmarking involves comparing HR metrics and practices against industry standards or best practices to identify areas for improvement.
- 5. Benefits of HR Analytics: ļµ Data-Driven Decision Making: HR analytics enables HR professionals to make decisions based on data and evidence, rather than intuition or guesswork. ļµ Improved Recruitment and Retention: By analyzing data on employee turnover, performance, and engagement, HR can identify factors that contribute to attrition and develop strategies to improve retention. ļµ Enhanced Employee Experience: HR analytics can help HR professionals understand the needs and preferences of employees, enabling them to personalize the employee experience and improve satisfaction. ļµ Cost Savings: By optimizing HR processes and improving workforce productivity, HR analytics can lead to cost savings for the organization. ļµ Strategic Alignment: HR analytics can help align HR initiatives with overall business goals, ensuring that HR strategies contribute to the organization's success.
- 6. Applications of HR Analytics: ļµ Recruitment and Selection: HR analytics can help HR professionals optimize recruitment processes, identify the best sources for talent, and improve the selection of candidates. ļµ Performance Management: HR analytics can help HR professionals evaluate employee performance, identify top performers, and develop strategies to improve performance. ļµ Employee Engagement: HR analytics can help HR professionals measure employee engagement, identify drivers of engagement, and develop initiatives to improve engagement levels. ļµ Retention: HR analytics can help HR professionals identify employees at risk of leaving the organization and develop retention strategies to keep them engaged. ļµ Workforce Planning: HR analytics can help HR professionals forecast future workforce needs, identify skill gaps, and develop strategies to address them. ļµ Diversity and Inclusion: HR analytics can help HR professionals measure diversity and inclusion within the organization, identify areas for improvement, and develop strategies to promote diversity and inclusion.
- 7. Types of HR Analytics ļµ Descriptive Analytics: ļµ Descriptive analytics involves analyzing historical data to understand past trends and patterns in HR metrics. It provides a summary of what has happened in the past, such as employee turnover rates, recruitment metrics, and training effectiveness. ļµ Predictive Analytics: ļµ Predictive analytics uses statistical algorithms and machine learning techniques to forecast future trends and outcomes based on historical data. This type of analysis can be used to predict employee turnover, identify high-potential employees, and forecast workforce needs.
- 8. ļµ Prescriptive Analytics: ļµ Prescriptive analytics goes beyond predicting future outcomes and recommends actions to achieve desired outcomes. It uses optimization and simulation techniques to suggest the best course of action based on the predicted outcomes. For example, prescriptive analytics can recommend strategies to improve employee engagement based on predictive models. ļµ Diagnostic Analytics: ļµ Diagnostic analytics involves analyzing data to understand the root causes of specific HR issues or trends. It helps HR professionals identify factors contributing to problems such as high turnover rates or low employee engagement, enabling them to take corrective actions. ļµ Workforce Planning Analytics: ļµ Workforce planning analytics involves analyzing data to forecast future workforce needs and develop strategies to address them. It helps HR professionals understand the current workforce demographics, skills gaps, and future hiring needs.
- 9. ļµ Talent Acquisition Analytics: ļµ Talent acquisition analytics focuses on analyzing data related to recruitment and hiring processes. It includes metrics such as time-to-fill, cost-per-hire, quality of hire, and source of hire, helping HR professionals optimize their recruitment strategies. ļµ Learning and Development Analytics: ļµ Learning and development analytics involves analyzing data related to training and development programs. It helps HR professionals assess the effectiveness of training programs, identify skill gaps, and allocate resources more efficiently. ļµ Employee Engagement Analytics: ļµ Employee engagement analytics focuses on analyzing data related to employee engagement, satisfaction, and retention. It helps HR professionals understand the drivers of engagement and develop strategies to improve employee morale and retention.
- 10. HR analytics ā¦ā¦ enables an organization to measure the impact of a range of HR metrics on overall business performance and make decisions based on data. ā¦ā¦ provides data-backed insight on what is working well in human capital assets and what is notā¦ so that organizations can make improvements and plan more effectively for the future. To me HR Analytics is of great importance in an organization because it enhances employee well being and helps keep them happy and motivated in their work.
- 11. HR Analytics is a multidisciplinary approach to integrate analytical and quantitative methodology into HR decision making. The objective is to improve the quality of people-related decisions, in order to improve individual and organizational performance. HR analytics plays a role in every aspect of the HR function, including recruiting, learning/training and development, engagement, retention, compensation, attrition and benefits.
- 12. Evolution: The evolution of HR analytics has come a long way since it was introduced. Organizations today have the ability to track all kinds of metrics to ensure that their people analytics, talent analytics, and workforce analytics are effective and accurate. Traditionally organizations made use of: ā¢ HR Metrics: Specific measures of HR practices used in the past , which led to.. ā¢ HR Analytics: An evidence-based approach to making HR decisions on the basis of analytical tools and models which are predictive and futuristic. The analytical tools are now used in the context of Machine learning too.
- 13. Key HR Metrics
- 14. So, from the metrics- way we have now come to applying HR Analytics, which comprises : ļµ A methodology for developing innovative insights ļµ Linking HR data, survey data, and business or organization data (financial, client, operationalā¦) ļµ Applying analytical and machine language techniques ( Factor and Cluster analysis, correlations, regression, Logistic regression, Classification trees modeling, using natural language programming, text mining, word cloudsā¦); ļµ Quantifying (or identifying) the impact of investments in human capital on the business or organizationā. [4 points Definition (the emphasis is on Impact)]
- 15. In todayās world advancement of technologies has combined with predictive analytics and has exponentially enhanced HR purposes in every HR aspect. HRPA ( HR Predictive Analytics) generates insights that cannot be achieved through traditional Benchmarking as the traditional analytics is reactive whereas HRPA is proactive and helps to predict the future using statistical, machine learning and data mining modeling systems. The benefits of HR analytics are many including, improved ROI, improved retention rates, and improved business processes.
- 16. Specifying Benefits of HR analytics today HR analytics, thanks to the advances in technology, have come a long way since they were first introduced. From being able to only evaluate goals and KPIs on data-gathering, thanks to the evolution of HR analytics, HR professionals can do so much more with their data-driven metrics. Advances in technology would mean data collection and storing facilities, computer capabilities to analyze data softwares available to apply on the data.
- 17. Examples for Benefits of HR Analytics: Recruiting employees who will be high performers. Organizations have a huge amount of data about their employees: their personal details, education, family background and so on. They also have data about employeesā performance and behaviors. Hence, they can correlate all these dimensions to determine the typical profile of an employee who is likely to be successful with them. Thatās the power of predictive analytics!ā Employees Leaving the Company One of the many benefits of HR analytics today is the ability to properly determine the likelihood of any employee leaving the company. This HR metric is predictive based on specific variables and can save the company a lot of money when it comes to retention and recruitment efforts.
- 18. Examples of Benefits of HR Analytics Employee Engagement Employee engagement has become a very important HR metric for organizations today as more and more businesses begin to realize the impact engaged employees have on their bottom line. Measuring employee engagement allows employers to know the extent to which employees are involved in different projects, allows employees to identify with the corporate image or brand and believe in the organization, as well as allows employers and employees to align with the strategy of the business. The benefit of this new HR metric is improved engagement, retention and reduced turnover rates.
- 19. Benefits of HR Analytics: Performance Prediction Another benefit of HR analytics today is being able to hire employees based on what they can bring to the company, rather than what their background is. Many employers today have begun to realize that having a degree doesnāt necessarily make a candidate a better match for an open position and in some cases, their experience and soft skills such as attitude towards work ethics and ability to operate in teams, are much more important. Thanks to the evolution of HR analytics, companies today can ensure that theyāre hiring the best employees to fill any open positions, based on the expected performance they would bring to the table.
- 20. From being able to only gather data but not do much with it, to being able to determine the likelihood of an employee staying with your company for an extended period of time, the evolution of HR analytics is still an ongoing process but continues to improve HR processes at all types of organizations. Without proper HR Analytics implemented at your organization, you will be able to experience the massive amount of benefits of HR analytics today. Source: Evolution of HR Analytics: Everything You Need To Know https://www.techfunnel.com/hr-tech/evolution-of-hr-analytics-everything-you-need-to-know/ By Emily Pribanic - Last Updated on February 17, 2020
- 21. Recruitment Learning & Development Engagement Performance Attrition Looking at HR Value Chain framework
- 22. Recruitment Phase. Recruitment refers to the overall process of identifying, attracting, screening, shortlisting, and interviewing suitable candidates for jobs (either permanent or temporary) within an organization. Traditionally, recruitment and selection used to define the process when people would apply for jobs using traditional paper applications and resumes, compared to people who apply online. Companies used traditional interviewing and hiring techniques and make a selection for candidates who would be expected to do well in the organization. In fact the selected candidates would be expected to bring significant value to the company.
- 23. Traditional recruitment methods have and are still being utilized by employers across the globe. The simplicity and familiarity of using these methods such as; paper-based job postings, internal hiring, referrals and word-of-mouth, are what make these still popular choices amongst lots of hiring professionals today. However, these techniques within hiring and recruitment are simply not enough to acquire top talent in a technological era. Instead, a more modern (& slightly updated) approach means hiring teams can make more effective and efficient hiring decisions with the assistance of technology.
- 24. Analytical Models we use in Recruitment: For Recruitment phase : We could use a simple approach to scoring of candidates, where Score can be a weighted average of skills which are required. The weights Can be determined by importances attached to the different skills, as obtained from the managers and supervisors. This could be carried out with use of Analysis of Variance as well as Multiple Linear Regression to determine what factors lead to possible high performance. Many softwares including excel and R- Studio can do this.
- 25. Learning and development Learning and development is a systematic process to enhance an employeeās skills, knowledge, and competency, resulting in better performance in a work setting. Specifically, learning is concerned with the acquisition of knowledge, skills, and attitudes. Development is the broadening and deepening of knowledge in line with oneās development goals. Learning, training, and development are often used interchangeably. However, there are subtle differences between these concepts:
- 27. Analytical Models we may use in the L&D phase: Skills Gap Analysis. An effective learning and development strategy relies on a process in which one continually moves through learning, attaining competencies and then Move to organizational capabilities. An essential part of this process is to determine the competencies/skills which are required in the organization and the level of these competencies already in existence with the employees. This is the Skills Gap Analysis as applied to the Learning & Development phase.
- 28. To give an example: Consider a list of competencies required in employees. When an organization looks at competencies required in employees, it often considers the following skills : Communication skill, written and spoken Business Acumen. Business Acumen is keenness and quickness in understanding and dealing with a business situation in a manner that is likely to lead to a good outcome Planning & Organization, refers to the skill of steps to be taken, prioritize tasks and time management Problem solving skills , This refers to defining a problem; determining the cause of the problem; identifying, prioritizing, selecting alternatives for a solution; implementing a solution. Future Developments, process that helps the company establish and maintain relationships with future prospects, learn about changing markets, supply chain alterations and changing macro economic environments
- 29. Changing Business Landscapes : responding to the new global challenges , changing supply chains, changing labor standards, human rights, climate change, market place Process and performance indicators, specifically focus on business processes measuring and monitoring operations, HR, Marketing and supply chain processes Identify Resource Requirements, to provide services, reach markets, maintain relationships with clients and earn revenue. Supply chain management (SCM) is management of the flow of goods, data, and finances related to a product or service, from the procurement of raw materials to the delivery of the product at its final destination Market knowledge :to know about the potential consumers behaviour which is directly and indirectly connected to the products and services
- 30. One can apply skills gap analysis in the L& D phase to identify which skills Need to be enhanced by training ( L&D) communication businessacumen planning&organization problemsolvngskill futuredev changing business landscape processes and performance indicators identify resource requirements supplychain market -10 0 10 Skills Gap Analysis reqd exist
- 31. Analytical Models we can apply L&D phase: Principal Component Analysis: To determine inherent dimensions in the skills required in the employees To determine inherent dimensions existing in the employee skill set. K-means Cluster Analysis to determine groups of employees having homogenous skills.
- 32. Factor Analysis in HR context: Principal Component Analysis Assume there are many variables under consideration many of which are interrelated. So there are groups of inter correlated variables. Each of these groups is represented by a Factor or Dimension, F1, F2 ā¦ Factors are linear combinations of variables so that F1 = a11V1+a12V2+ā¦ F2 = a21V1+a22V2+ā¦
- 33. Consider an example in Factor Analysis: Consider a database having 6 competencies on which Employees have been rated. So there are 6 variables: V1= competency on business acumen V2= competency on planning & development V3= competency on problem solving skill V4= understanding future development V5= understanding on changing lanscape V6= competency on communication
- 34. The six variables have the following correlation martrix between them: v1 v2 v3 v4 v5 v6 ------------------------------------------------------------------------------------------- V1 1 .8 .9 .1 .06 .08 V2 1 .85 .08 .1 .04 V3 1 .02 .05 .04 V4 1 .9 .08 V5 1 .02 V6 1 ------------------------------------------------------------------------------------------- There is very high correlation between v1,v2 and v3. Also very high correlation between v4 and v5. v6 is on its own. This result indicates that there are three factors,F1,F2 , F3 ; F1 representing v1,v2 and v3. F2 representing v4 and v5. F3=v6 F1 measures a dimension which is common to v1,v2,v3. F2 measures another dimension, common to v4,v5 and F3 measuringv6.
- 35. F1: can be labelled knowledge of business sense F2: changing business scenario F3: communication Rotation of factors : To help clearer identification of factors we make use of varimax rotation. Other approaches are quartimax, equamax. Save factor scores for further analysis
- 36. Concepts of cluster analysis in HR Typically, Cluster Analysis applications are observed in market segmentation studies. This multivariate approach identifies homogeneous groups, called Clusters of respondents in a way that the members in any cluster are similar to each other but different from the respondents in the other clusters. In our HR context we often apply Cluster analysis to identify groups of employees with homogeneous skills.
- 37. In the Learning & Development phase, an organization often needs to conduct a Training Evaluation process: The Kirkpatrick Model is a globally recognized method of evaluating the results of training and learning programs. It assesses both formal and informal training methods and rates them against four levels of criteria: reaction, learning, behavior, and results.
- 38. Level 1: Reaction The first level of criteria is āreaction,ā which measures whether learners find the training engaging, favorable, and relevant to their jobs. This level is most commonly assessed by an after-training survey (often referred to as a āsmile sheetā) that asks students to rate their experience. Level 2: Learning Level 2 gauges the learning of each participant based on whether learners acquire the intended knowledge, skills, attitude, confidence and commitment to the training. Methods of assessment include exams or interview-style evaluations. A defined, clear scoring process must be determined in advance to reduce inconsistencies. Level 3: Behavior One of the most crucial steps in the Kirkpatrick Model, Level 3 measures whether participants were truly impacted by the learning and if theyāre applying what they learn. Assessing behavioral changes makes it possible to know not only whether the skills were understood, but if it's logistically possible to use the skills in the workplace. Level 4: Results The final level, Level 4, is dedicated to measuring direct results. Level Four measures the learning against an organization's business outcomesā the Key Performance Indicators that were established before learning was initiated. Common KPIās include higher return on investments, less workplace accidents, and larger quantity of sales. https://www.ardentlearning.com/blog/what-is-the-kirkpatrick-model
- 39. Level 5 in Kirkpatrick Model Level 5 isnāt featured in the original Kirkpatrick Model but is an additional level of evaluation created by Jack Phillips that uses cost-benefit analysis to determine the value of training programs. In other words, it helps companies calculate whether the money they spent on training produced measurable business results. The standard formula for measuring the ROI Of training is as follows: ROI (percentage) = ((Monetary benefits ā Training costs)/Training Costs) x 100.
- 40. Learning and Development has a dilemma too: To train or not to train.
- 41. Engagement: Employee engagement can be explained as a positive employeesā emotional attachment and Commitment to the organization. Robinson et al. (2004) define employee engagement as āa positive attitude held by the employee towards the organization and its value. An engaged employee is aware of business context, and works with colleagues to improve performance within the job for the benefit of the organization.
- 42. Engagement is about passion and commitment-the willingness to invest oneself and expand oneās discretionary effort to help the employer succeed, which is beyond simple satisfaction with the employment arrangement or basic loyalty to the employer (BlessingWhite, 2008; Erickson, 2005; Macey and Schnieder ,2008). There are four key drivers of engagement: Job Manager People Organization Employees can become disengaged where thereās an imbalance or misalignment in any of these areas.
- 43. Employee Engagement is key to success in any organization . There is a clear link between engaged employees and the level of customer service that company can provide and bottom line. Lack of employee engagement leads to organization culture being detrimental to physical and mental health of employees. Substandard nature of the job, discontented personal factors, uncongenial organizational support, dispirited perceptual factors and hostile organizational culture are the main determinants of lack of engagement and attrition in the organization.
- 44. The Employee Engagement Score ā A Simple Way to Calculate Engagement The most common measure of Employee Engagement is to create a Employee Engagement Score through an engagement survey. This can be supplemented by other indicators that are proxies for engagement. This includes absence and attrition rates, as well as indirect factors such as Emplo Satisfaction (ESAT) or Net Promoter Score, NPS (measures employee experience and predicts business growth) . In addition, there is no substitute for walking round the company and talking with people. Good leaders will instinctively have a good feel for the mood of the company and how engaged everyone is
- 45. The third phase: Engagement Engagement is the state of emotional and intellectual involvement an employee has to an organization. Organization encourages employees to say, stay and strive. Strive Harder and Contribute Encourages employees Stay Committed and Loyal Encourages employees to be Brand Ambassadors ā Create positive environment ā Attract and retain ā Talent ā Customers ā Business Partners ā Employee retention ā Continuity and consistency ā Engaged Employees drive growth ā Engage Customers ā Improve Productiivity ā Provide sustainable competitive advantage 1 Say Say 1 Say Say Stay Stay 2 Stay Stay 2 3 Strive Strive 3 Strive Strive
- 46. Employee engagement score: I would not hesitate to recommend this company to a friend seeking employment Given the opportunity, I tell others great things about working here It would take a lot to get me to leave this company I rarely think about leaving this company to work somewhere else This company inspires me to do my best work every day People here are always willing to give effort beyond requirements to help the organization succeed SAY STAY STRIVE India Best Employers 82 % 91 % 86 % 88 % 79 % 91 % Company Scores 71 % 78 % 69 % 68% 57 % 84 % ā¢ High āSAYā scores reflective of a strong employer brand ā¢ STAY scores lower than 60% would be an indication of talent risk ā¢ āSTRIVEā scores reflective of a workforce willing to go ābeyond the briefā The Employee Engagement score was determined by responses to the following questions:
- 47. Engagement Survey : Employee Engagement is often measured in terms of different attributes which have been identified with a lot of research. These are: Perceived value of job The employee should see the following attributes in his/her job: *High Perceived Value of job content ā¢ presence of Equality ā¢ presence of Advancement Opportunities ā¢ No Incompatible Policies Teamwork environment Organizational Culture *Do not see Power and Politics ā¢ satisfied with Colleagues and teamwork *culture of promoting/supporting teamwork
- 48. Compensation Personal Factors ā¢ Satisfied with compensation ā¢ Compensation comparable with other organizations ā¢ OK with leave and working hours Positive Nature of Job ā¢ Presence of Skill Variety among colleagues ā¢ Interesting Nature of Job ā¢ Presence of Challenge .Organizational Support ā¢ Regular Working Hours ā¢ Emphasis on quality as well quantity ā¢ Effective Supervision
- 49. Self-Fulfillment Factors * Lack of Partiality ā¢ Achievement always recognized * Good Mentoring ā¢ Sure of career growth Engagement I feel fully engaged on my job in my company: NA SomewhatA A stronglyagree
- 50. Multiple Regression : To find out what features to look for in candidates so They are high performers. Dependent variable : High-engagement, measured in 4-point Likert Scale Engagement Perceived nature of job Organizational support Teamwork environment Compensation Culture of appreciation
- 51. Performance Evaluation System Performance evaluation is the process of evaluating how effectively employees are fulfilling their job responsibilities and contributing to the accomplishment of organizational goals. To appraise performance effectively, a manager must be aware of the specific expectation for a job, monitor the employeeās behavior and results, compare the observed behavior and results to expectations and measure the match between them. The primary goals of a performance evaluation system are to provide an equitable measurement of an employeeās contribution to the workforce, produce accurate appraisal documentation to protect both the employee and employer, and obtain a high level of quality and quantity in the work produced.
- 52. Consider a performance evaluation being carried out over several departments in an organization. First it is necessary to determine the specific items on which the employees will be evaluated. Planning & organization Quality & accuracy Analytical aptitude Oral & written communication Productivity Client orientation Innovation as demonstrated Leadership exhibited
- 53. We consider next an analytical model: to determine attributes in Performance leading to ā promotionā. This could be conducted by Logistic Regression or Decision trees. Here we consider a binary categorical dependent variable called āpromoteā. In the database promote is = 1, if employee did get promoted, 0 otherwise. This is also called Target variable. The predictor variables comprise : several demographic variables, communication exhibited, leadership exhibited, Variety of skills exhibited, Client orientation exhibited, Initiative exhibitedā¦
- 54. A Machine Learning Approachā¦ Decision Trees Decision Tree approach, which is a combination of heuristic and statistical modeling, includes many algorithms which work efficiently in organizations having extensive data bases. Of these recursive partitioning algorithms, or decision trees, are a versatile tool for finding out trends, patterns or relationships in the data. There are many algorithms under the decision tree approach which contain procedures capable of combing through a large set of predictor variables , successively splitting a data set into subgroups to improve the prediction of the target dependent variable . Usually, decision tree approaches work very well when there are too many predictor variables. The algorithm starts with a target variable usually taking up values of 1 or 0 . It starts with a base node called root node. The root node gives the response rate as well as the total size of the sample being considered in the data base.
- 55. Classification Trees, CART: Classification and Regression Trees Here the dependent variable, also called target variable, is āpromoteā Predictor variables : Demographic variables such as age, income, eduā¦ Performance in Planning and Organizing Work, etc. The tree has circles and rectangles where rectangles are terminal variables, which only has branches coming in. The tree starts with most significant variable on top. The tree grows by Recursive Partitioning Steps.
- 56. Case: Determine attributes in Performance leading to ā promotionā Consider a large employee database in a company containing demographic variables And performance related attributes such as : Performance in planning and organization Performance related to Quality & Accuracy Communication demonstrated Leadership demonstrated Variety of skills demonstrated Client orientation demonstrated Initiative exhibited. These are the predictor variables. In the employee database there is also a column āpromoteā =1, if employee got Promoted in last 5 years, 0 otherwise. This is the Target variable. Also called root node.
- 57. Classification Tree by SPSS
- 58. Classification Tree by R-Studio
- 59. Employee Attrition Employee retention is a key objective for most HR organizations, and employee turnover is the single most concern in most prevalent HR metric. However, knowing your turnover rate does little to support strategic business plans. To achieve true insight, a more in-depth analysis of whatās causing turnover in different parts of the organization is required. In order to reduce employee turnover, HR needs to become more data-driven, looking past simple descriptive analytics and towards more exploratory analytics, and (even better) predictive analytics. If the organization gets to know which employees are high risk for attrition, preventive initiatives can be taken to address the concerns of the employee who is leaving. HR Analytics helps to determine the probability that a certain employee may like leave.
- 60. Predictive Attrition Model ā Logistic Regression Through predictive algorithms, companies gain better understanding and can undertake preventive measures for employee attrition. On a basic level, the model works by classifying employee profiles based on various attributes such as age, sex, marital status, education level, work experience, distance from hometown, etc. and generates various levels of risk of attrition. Occasionally, other parameters like performance over the years, pay raise, work batch, educational institution are also taken into consideration. However, the accuracy of the model is directly proportional to the selection of parameters, which in turn, leads to the generation of the ātypeā of predictive model most suitable for the organization.
- 61. Creating the Model Various statistical and machine learning algorithms are designed to construct the predictive models to determine factors leading to attrition. We can use the Logistic Regression model which gives the āprobability of attritionā when the outcomes are dichotomous, stay or leave. Consider an employee database where employee demographic variables are available, academic background, previous experience, if workplace is far from home, state of health ( self assessed), size of previous company worked, if previous company was MNC , perceived growth opportunities in the current Company, perceived culture of appreciation in the current company, level/amount of training received in the company, salary benefit and whether the employee stayed or left.
- 62. Analyzing employee feedback : When unexpected attrition happens the company must find a way not just to see how many people were leaving but to understand why they were leaving, if employees would continue to leave, and how to adapt the workforce strategy to prevent employees leaving. One method would be to take exit feedback. This could be in terms of a quantitative survey similar to previous case. Or in terms of qualitative feedback, in terms of document, word file for example. We have discussed how to analyze quantitative survey. Now we discuss how to analyze documentsā¦ with the help of text mining.
- 63. Text Mining Process Text Mining has wide applications in: Customer relationship Management, analyzing customer complaints customer preferences, Market Analysis, E-Commerce, analyzing online buying behavior, and Natural Language Processing, NLP, an application in Artificial Intelligence, relating to interaction between computers and languages used by humans Text mining deals with converting : Unstructured data ( not in tables) to numeric data to be used for analysis, using frequency tables, association analysis and Word Clouds.
- 64. Text Mining process: Step 1 : read file, usually in Plain Text format Step 2 : Build Corpus, A corpus is a collection of texts, written or spoken, usually stored in a computer database. Step 3: Clean text, remove punctuations, uppercase letters, common English words. Step 4, Create Term Document Matrix, TDM, Matrix for words and frequency. Step 5: Create tables, frequency distributions, Correlations, Word Clouds Step 6: Interpret findings and derive patterns and insights.
- 65. Create a Word Cloud : Attrition is mostly related to credit ( or lack of it) given to employeesāwork in the organization.