Scribd
Upload
66 views7 pages

Workforce Analytics for Recruitment Success

Mr. Suresh, head of HR at Alvin Technologies, observed that 30-40% of candidates offered jobs eventually rejected them. Niraj, an employee in Alvin's HR analytics program, analyzed historical data of 7,000 candidates using machine learning to predict which variables affected joining and by what odds. His logistic regression model accurately predicted joins and non-joins. He then predicted which recently interviewed candidates were most likely to join based on the model.

Uploaded by

Mohan Krishna
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
66 views7 pages

Workforce Analytics for Recruitment Success

Mr. Suresh, head of HR at Alvin Technologies, observed that 30-40% of candidates offered jobs eventually rejected them. Niraj, an employee in Alvin's HR analytics program, analyzed historical data of 7,000 candidates using machine learning to predict which variables affected joining and by what odds. His logistic regression model accurately predicted joins and non-joins. He then predicted which recently interviewed candidates were most likely to join based on the model.

Uploaded by

Mohan Krishna
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd

ASSIGNMENT 2: WORKFORCE ANALYTICS

Alvin Technologies, a large software development and project management company located
in Southern India recruits nearly 2000 engineers with specialization in Computer Sciences or
Information Technology from the industry. Every year their need for manpower was on the
rise. In the Indian software job market, number of engineers was more, but quality of most
engineers was not up to the mark as set by the top management of Alvin Technologies.
Hence, the recruitment screening process was conducted meticulously to identify the
candidates with the right potential and after a rigorous five-round interview process, the best
of the lot were extended offer letters.
The HR Head of Alvin, Mr. Suresh KG observed that there was trend of roughly 30-40% of
the candidates who were given joining offer to reject the same for better opportunities or
reasons unknown. This was a serious worry for Mr. Suresh, since the cost and effort put in to
zero down onto the final candidates was substantial. He was left pondering why so many
individuals decide to not be part of Alvin, despite the firm taking care of major exigencies for
the new joinees such as relocation expense coverage and flexible joining time to avoid notice
period blues. Alvin was considered a good pay master and hence Mr. Suresh could not find
any satisfactory explanation to this pattern. One of the junior staffs in the HR department
-Mr. Niraj Babu approached the Head, HR and suggested that he may throw some light on the
matter. Niraj has been enrolled in a HR Analytics certification course at a premier B School
in India for past three months. This program was sponsored by Alvin, and Niraj felt obliged
to help his department head with the pressing issue by sharing some analytic insights that he
has picked up in his course. Niraj sought permission to get access to the existing data set of
roughly 7000 employees who have appeared for interview till the offer letter stages for the
firm in the past three years. He then first cleaned the existing data for any anomalies such as
missing values and outliers. Finally, he fit the data in a ML model to understand the pattern
behind the event of joining / not joining. Lastly, he applied the trained model on a data set of
new candidates who were about to be extended offer letter by the firm.

QUESTIONS

1) Conduct a predictive analysis to test which of the independent variables are affecting the
chance of joining for the candidates and by what odds with the Training data.
2) Develop a ML model with the Training data using Logistic Regression to test the accuracy
with which the outcome (joined / not joined) is getting predicted in terms of precision, recall,
specificity and AUC.
3) Predict which candidates are most likely to join based on the ML model with more
accuracy as found using the fresh data (Test data) of recently interviewed candidates.
Answer
As we know if odds ratio is >1.1 then that variable is affecting the chance of joining.

So here we can see

Notice period has odds of 1.4 i.e., chances of candidate joining 1.4 times more if notice period is
there

Relocation required odds are 51.7 i.e., chances of candidate joining 51.7 times more if candidate is
offered relocation

Joining bonus odds are 0.62 <0.9 i.e., chances of candidate not joining 0.62 if joining bonus is given

Rem factors have odds ratio between 0.9 & 1.1 so they can be omitted or can be concluded that they
are not affecting candidates joining

Joining Bonus

Ans 2
Confusion Matrix

ROC Analysis

Rank
Test And score-
Average

Ans 3
Prediction
People most likely to join
Most likely to not join

There are high chances almost all will join except 22 people.

Common questions

Powered by AI

HR analytics played a transformative role by revealing data-backed insights into factors influencing candidate decisions, such as the high effectiveness of relocation packages and the limited impact of joining bonuses. This information allowed Alvin Technologies to strategically refine their hiring process, focusing on offerings and conditions that most positively affect acceptance rates. By integrating analytics into their recruitment strategy, Alvin could make evidence-based decisions to better attract and retain desired talent .

The main obstacles identified included the significant drop-off rate of 30-40% for candidates who were extended offers and the exhaustive five-round recruitment process that still failed to assure commitment. Analytics can help overcome these obstacles by identifying key factors that influence joining decisions, such as relocation offer effectiveness and the counterproductive nature of joining bonuses. By leveraging these insights, Alvin Technologies can optimize their recruitment strategies to target high-probability candidates and make informed decisions on incentivization, thereby reducing drop-off rates and enhancing process efficiency .

To address the candidate dropout issue, Alvin Technologies' HR department, led by junior staff member Niraj Babu, utilized data analytics by accessing a dataset of interviews and offer letters from the past three years. Niraj cleaned the data to remove anomalies and used a machine learning model to analyze patterns influencing candidate joining decisions. The analysis revealed that relocation offers significantly increased the likelihood of candidates joining, with an odds ratio of 51.7. In contrast, offering a joining bonus decreased the likelihood, indicating its ineffectiveness in attracting candidates .

The odds ratios suggest several conclusions for Alvin Technologies. A notice period had an odds ratio of 1.4, indicating a slight positive influence on candidate decisions to join if it is present. The relocation requirement had a strikingly high odds ratio of 51.7, making it a highly effective incentive for candidates to accept an offer. Conversely, the joining bonus had an odds ratio of 0.62, suggesting it was counterproductive, as it decreased the likelihood of candidates joining. These insights should guide Alvin in refining its HR policies to better attract qualified candidates .

Niraj Babu employed a Logistic Regression model to predict the probabilities of candidates joining Alvin Technologies. To evaluate the model's performance, he used various metrics such as precision, recall, specificity, and the Area Under the Curve (AUC) of the Receiver Operating Characteristic (ROC) analysis. These metrics helped to assess the accuracy and reliability of the predictions made by the model .

To improve offer acceptance rates, Alvin Technologies should consider emphasizing and potentially expanding their relocation package offers, given its high influence on positive candidate decisions with an odds ratio of 51.7. They should also reconsider the effectiveness of joining bonuses, which had a negative effect on the likelihood of acceptance with an odds ratio of 0.62, suggesting limited or no such incentives might be more effective. Additionally, evaluating whether the notice period can be more flexibly managed might further enhance acceptance rates .

The odds ratio analysis provided counterintuitive insights into candidate incentives, challenging the common assumption that joining bonuses automatically enhance job offer attractiveness. Contrary to this belief, the analysis showed joining bonuses diminished the likelihood of candidates accepting an offer with odds ratio 0.62, suggesting they might have been perceived negatively or as insufficient. Meanwhile, a strong positive correlation with relocation packages was highlighted with an odds ratio of 51.7, directing attention to their significant impact. These revelations pushed Alvin Technologies to reconsider and realign their incentive strategies to align with actual candidate preferences and behaviors .

Niraj Babu started by obtaining permissions to access a historical dataset and then focused on cleaning the data to eliminate anomalies such as missing values and outliers, a crucial step to ensure data quality for analysis. He then implemented a machine learning model, specifically logistic regression, to identify patterns and predict outcomes based on the cleaned data. This methodological approach allowed a systematic and reliable means of discerning the factors influencing candidate decisions to join and applying these insights to predict future behaviors effectively .

The data analysis indicated that offering a relocation package had a profoundly positive influence on candidates' decisions to join Alvin Technologies, with an odds ratio of 51.7. This suggests that candidates were 51.7 times more likely to join the company if they were offered a relocation package, highlighting the package's significant effectiveness as a motivating factor for candidates considering the offer .

Alvin Technologies faced a challenge in recruiting engineers despite having a high demand for them annually. The primary issue was the mismatch between the quantity and quality of engineers available in the Indian job market, as most candidates did not meet the company's high standards. Additionally, there was a significant drop-off rate, with 30-40% of offered candidates declining to join the company, leading to concerns about the effectiveness and cost of the recruitment process. The company attempted to address these challenges by implementing a meticulous and rigorous five-round interview process to select the best candidates .

You might also like