Written by: Team PredictiveHire
An HR Algorithm Can Tell How Often You Will Change Jobs
Job-hopping algorithm: Assessing Job-Hopping Attitudes From Chat Interview
We have no survivable and sustainable future without science, just as we do not have you without it.
Since the start of the coronavirus epidemic, many companies have turned to smart algorithms to find out who is the best candidate for open positions. Most often, face-finding programs, games, quizzes, and software that examines other visual or linguistic patterns are used to decide who is included in the interview circle.
An Australian company called Sapia (Formerly PredictiveHire), founded in October 2013, appears to have gone much further. It has developed a machine-learning algorithm to assess the likelihood of frequent job changes for a given candidate. – MIT Technology Review.
According to Barbara Hyman, CEO of HR, their clients are employers who have to process a lot of application. Also they are active in the areas of customer service, retail, sales or healthcare, among others.
In the first round, a chatbot decides on the applicants
When someone applies for a job through an HR company, they must first “convince” a chatbot of their values. The algorithm asks a series of open-ended questions and analyses personality traits such as initiative, intrinsic motivation, or resilience.
Moreover, the algorithm may examine the likelihood of frequent job changes in the future – or, as advertised on the Sapia website, the “ risk of escape ” – even for fully career candidates. The focus of the HR company’s latest study is to develop a machine learning algorithm that specifically seeks to predict this. The research examined 45,899 candidates. They had previously answered 5-7 open-ended questions about their experiences and situational awareness through the Sapia chatbot.
The chatbot asked for personality traits that, based on Sapia’s own research, may be closely related to frequent job changes. For example the traits could be -greater openness to new experiences or lack of practicality.
Algorithms against wage increases
Nathan Newman, an associate professor at John Jay College of Criminal Justice in New York who wrote a study in 2017 on how large-sample data analysis can be used to break wages in addition to discriminating against employees, told MIT Technology Review Recent work by Sapia.
This includes the increasingly popular personality tests based on machine learning, which seek to screen out potential workers who are more likely to support unionisation or are more likely to ask for wage increases. According to MIT Technology Review, employers are increasingly keeping an eye on their employees ’emails, online chats, and data they can use to filter out whether a particular colleague is about to leave. All this so they can calculate the minimum wage increase is and where appropriate, they may be allowed to remain.
Uber’s algorithm-based management systems are said to seek to keep employees away from offices and digital locations in a way that they can’t even accidentally organize and collectively demand better pay or treatment.
Sapia has found a relationship between the language people use and their attitudes towards job-hopping.
If a simple automated chat interview can infer a candidate’s likelihood of job-hopping, it presents significant opportunities, especially when assessing candidates with no prior work history.
This work shows that the language one uses when responding to interview questions related to situational judgment and past behaviour is predictive of their likelihood to job hop. This paper explores:
- Research around self-initiated job hopping
- Correlation between language and job-hopping likelihood
- NLP methods that can be used to represent language
To keep up to date on all things “Hiring with Ai” subscribe to our blog!
Also, have you seen the 2020 Candidate Experience Playbook? Download it here.