We all hear about the importance of “socio-emotional skills” when looking for a job. Employers are said to be looking for individuals who are hardworking, meet deadlines, are reliable, creative, collaborative … the list goes on depending on the occupation. In recent years, it seems, these skills have become equally important as technical skills. But do employers really care about these soft skills when hiring? If so, what type of personality do they favor?
To answer these questions, the World Bank partnered with an online job portal in India called Babajob (acquired by Quickr Jobs in June 2017) to conduct a Randomized Control Trial (RCT) experimentation directly on their web portal leveraging their big data, including1.29 million job ads posted by 524,000 employers for 8 million registered job seekers.
We asked their job applicants to take a Big-Five type personality test using Babajob’s chat message software called “chatbot”. An 11-item version of the Big Five inventory was used: 3 questions on agreeableness, 2 questions on conscientiousness, 2 questions on openness to experience, 2 questions on emotional stability (neuroticism), and 2 questions on extraversion. The survey link was sent out via SMS messages through a chatbot. Applicants responded to a survey voluntarily. After job seekers completed the assessment, they got to see their scored results immediately. The results of the test were also automatically uploaded to their Babajob profiles, and were shown on their applications to various jobs if they were in the treatment group of the experimentation.
Outcomes of this experiment were measured by whether a potential employer assesses a seeker by opening (unlocking) his/her application and background information. We compared outcomes between those for whom the personality test results were shown and not shown to analyze the impact of non-cognitive skills on hiring.
Results
A total of 2,870 applicants filled the survey, of which 51.5% constituted the treatment group. We observed that the disclosure of personality test results (treatment) had a positive and quite significant impact on the number of employers clicking the profile of potential workers to show their interest. We also found a significant impact on the individuals who identified themselves as organized, calm, imaginative and/or quiet applicants. There was no effect on those who described themselves as easy-going, sensitive, realistic and/or outgoing applicants, which seems to indicate (typical) employers’ preference. We observed small differences between men and women but this finding is not very conclusive. Another interesting observation is the fact that some job categories (such as office clerk, sales, receptionist) seem to require non-cognitive skills more than other professions (at least, from the employers’ viewpoint).
Conclusion
The results show that socio-emotional skills play potentially important roles in the hiring process and that employers seem to have certain (stereotypical) preference when looking for employees. In other words, information on noncognitive skills is usually quite private, thus, employers generally do not have such knowledge on job applicants and may have to imperfectly guess their preference from available data. Thus, provision of socio-emotional skills information can be a significant source of improving efficiencies in the labor market.
Furthermore, this innovative experiment using Big Data by directly introducing experimentation on the online job portal adds value to the growing volume of big data research. Using big data from such online job portal has a low marginal cost on extra information acquisition and the accurate results become available right away. We can also easily test and adjust the parameters of the experiment if needed to improve the accuracy of the design and can easily collect the data overtime as it will be automatically stored in the portal. Given the interesting outcomes of this first experiment, we hope to leverage the potential of big data in labor market analytics more in the future.
For more details of this experiment, please access our publication: Asymmetric Information on Noncognitive Skills in the Indian Labor Market: An Experiment in Online Job Portal.
To answer these questions, the World Bank partnered with an online job portal in India called Babajob (acquired by Quickr Jobs in June 2017) to conduct a Randomized Control Trial (RCT) experimentation directly on their web portal leveraging their big data, including1.29 million job ads posted by 524,000 employers for 8 million registered job seekers.
We asked their job applicants to take a Big-Five type personality test using Babajob’s chat message software called “chatbot”. An 11-item version of the Big Five inventory was used: 3 questions on agreeableness, 2 questions on conscientiousness, 2 questions on openness to experience, 2 questions on emotional stability (neuroticism), and 2 questions on extraversion. The survey link was sent out via SMS messages through a chatbot. Applicants responded to a survey voluntarily. After job seekers completed the assessment, they got to see their scored results immediately. The results of the test were also automatically uploaded to their Babajob profiles, and were shown on their applications to various jobs if they were in the treatment group of the experimentation.
Outcomes of this experiment were measured by whether a potential employer assesses a seeker by opening (unlocking) his/her application and background information. We compared outcomes between those for whom the personality test results were shown and not shown to analyze the impact of non-cognitive skills on hiring.
Results
A total of 2,870 applicants filled the survey, of which 51.5% constituted the treatment group. We observed that the disclosure of personality test results (treatment) had a positive and quite significant impact on the number of employers clicking the profile of potential workers to show their interest. We also found a significant impact on the individuals who identified themselves as organized, calm, imaginative and/or quiet applicants. There was no effect on those who described themselves as easy-going, sensitive, realistic and/or outgoing applicants, which seems to indicate (typical) employers’ preference. We observed small differences between men and women but this finding is not very conclusive. Another interesting observation is the fact that some job categories (such as office clerk, sales, receptionist) seem to require non-cognitive skills more than other professions (at least, from the employers’ viewpoint).
Conclusion
The results show that socio-emotional skills play potentially important roles in the hiring process and that employers seem to have certain (stereotypical) preference when looking for employees. In other words, information on noncognitive skills is usually quite private, thus, employers generally do not have such knowledge on job applicants and may have to imperfectly guess their preference from available data. Thus, provision of socio-emotional skills information can be a significant source of improving efficiencies in the labor market.
Furthermore, this innovative experiment using Big Data by directly introducing experimentation on the online job portal adds value to the growing volume of big data research. Using big data from such online job portal has a low marginal cost on extra information acquisition and the accurate results become available right away. We can also easily test and adjust the parameters of the experiment if needed to improve the accuracy of the design and can easily collect the data overtime as it will be automatically stored in the portal. Given the interesting outcomes of this first experiment, we hope to leverage the potential of big data in labor market analytics more in the future.
For more details of this experiment, please access our publication: Asymmetric Information on Noncognitive Skills in the Indian Labor Market: An Experiment in Online Job Portal.
Join the Conversation