Pratibha Screening and Training Tool

Female labor force participation in South Asia has seen a decline in the past two decades and although the Indian garment manufacturing industry is a major employer of women, they mostly occupy entry-level positions and rarely move up the ranks to supervisory roles. This hampers the career aspirations of women in the workforce and also contributes to creating a hostile workplace where women don’t feel safe or productive. We use machine learning algorithms to build a screening and training tool, called Pratibha, which has the potential to promote gender equity and empowerment in the workplace by reducing bias in the managerial screening process during recruitment while also measuring and remediating skill deficiencies in prospective candidates. This low-cost solution has the ability to not only increase the representation of women in managerial roles but also improve the work environment for frontline workers, the majority of whom are women, through better-trained managers.

Our solution addresses the challenge’s focus on increasing the participation of underserved communities (especially women) in the creation of new jobs and opportunities. Through our tool, Pratibha, we seek to increase the opportunities for women to advance to supervisory roles in low-skill, mass-manufacturing contexts like the Indian garment sector.

Despite comprising a large share of employment in the garments sector, women mostly occupy entry-level positions and rarely move up the ranks to supervisory roles.  In India, the garment sector employs 45 million people, more than 60% of whom are women. Our partner, Shahi Exports Pvt. Ltd., is India’s largest apparel manufacturer employing over 100,000 workers and mirrors these demographic patterns with 72% women, the vast majority of whom are frontline workers as only 30% of supervisors are women. This pattern is repeated even on the global scale where it is estimated that 68% of the workforce in apparel manufacturing is female, however a vast majority of the higher paid supervisory roles are occupied by men (BSR, 2017). These numbers indicate that even in industries such as apparel manufacturing that employ a large female workforce, women are not getting full opportunities to climb the ladder. 

A key factor that has contributed to the creation of this imbalance are systematically inaccurate beliefs about women’s managerial skills. Despite their existing little evidence to that effect, incorrect beliefs and social stigma often lead to employers ignoring qualified women when taking decisions on promotions. Our tool seeks to overcome these biases.

This tool is aimed at improving the lives of low-income female frontline workers in the Indian garment manufacturing sector. These women are usually young, unmarried migrants from eastern India who migrate to one of the three garment manufacturing clusters of India.

The explosive growth of the apparel manufacturing sector from the late-1990s onwards created a huge demand for cheap, unskilled labor in these manufacturing clusters, and women from states like Bihar, Jharkhand, Chhattisgarh, Odisha and UP flocked to meet the demand. Though the jobs are in the formal and semi-formal sector they are characterized by a great deal of insecurity, stress and uncertainty usually reminiscent of the informal sector. The factory floor is often a high-stress environment where male supervisors push a predominantly female workforce to meet increasingly ambitious production quotas. This systemic imbalance, disconnection, and environment of fear, manifests in the form of shouting, abuse, and harassment on the factory floor and that these, along with other issues related to the work environment, often go unreported and unresolved.

Studies have shown that the presence of women in managerial roles often motivates female frontline workers as well as improves the general work environment for them, thus leading to an increase in overall worker well-being and firm productivity. Through this tool we seek to increase the chances for these low-income migrant women to advance to managerial positions and in-turn help them improve the work environment on the factory floor.

The Indian textile industry accounts for 14% of the overall industrial production, contributes to approximately 30% of its total exports, adds 4% to India's GDP and is the second-largest source of employment after agriculture. In an era characterized by jobless growth, the apparel manufacturing industry has been an exception and currently employs over 45 million people across the nation, a majority of whom are women. Further, industry reports tend to agree that the future is going to be marked by an expansion in the workforce employed and growth in productivity.

However most of the jobs that have been created in this sector are characterised by low-pay, insecurity, high-stress and general precarity; additionally women, who make-up a majority of the workforce, face sexual harassment, verbal abuse and hostility in the workplace from their mostly male supervisors. The gender imbalance amongst the supervisors contribute to the creation of a hostile workplace and also dampens the career aspirations of women.

Our solution will remove all bias from the screening process for new managers and also provide personalized vocational training to all those who require it. Thus, our tool Pratibha, would not only help increase the proportion of marginalized low-income women advancing to managerial roles but also improve the work environment for frontline workers, the majority of whom are women, through better-trained managers. By helping to make a safer and more inclusive workplace, Pratibha will also contribute to the development of an atmosphere that encourages more women to participate in the formal economy.

The team is lead by Dr Achyuta Adhvaryu, Chief Development Officer & Co-Founder of Good Business Lab (GBL). GBL is an India-based nonprofit organisation dedicated to improving the lives of low-income workers by proving that worker wellbeing - inside and outside the workplace - delivers measurable business returns to firms.

Our solution aims to address the limitations associated with supervisor hiring on multiple fronts - it empirically demonstrates which skills augment productivity, and how  these skills can be accurately measured. 

Our solution is also innovative in that we are looking to solve a nearly invisible gender imbalance that is not widely thought about. Over the past decade we have observed a decline in the already low share of women in factory-level supervisory roles.  Despite the fact that over 60% of the frontline workers are female, only about 30% of the supervisors are women. Upon deeper examination we realized that this is because men are more likely to get selected into interviewing for supervisory roles by those in-charge of hiring. We have found an innovative way to eliminate this biased selection.

Additionally, we developed our technology - the  Pratibha Screening and Training tool entirely from scratch in order to fit the specific circumstances of supervisors and factory workers. This level of ground-level input from all stakeholders makes this a unique and innovative tool in the mass-manufacturing context.

From a broad set of soft skills and managerial skills, we narrowed down to which specific ones best predict managerial productivity using surveys and machine learning techniques. We then developed a tablet based psychometric screening test - called Pratibha Screening - which is used to screen candidates for supervisory positions. The results of this test then get parsed through a proprietary predictive algorithm that ranks candidates on their productivity. 

The algorithm also generates a bespoke training path per candidate. The training path data is sent to another aspect of the tool we have developed - Pratibha Training - where we have created a tablet based individual training (we tested the curriculum for productivity and wellbeing impacts in an unpublished upcoming paper). The training tool can be accessed by a user who has completed the screening test. The access training content with a pre-quiz and post-quiz to gauge their levels of understanding, before and after taking the training. The entire interface has been gamified to motivate the user to complete the full training.

Our solution is designed to be scalable in that it works on minimally expensive equipment (like basic tablets and mobile phones), has content that can be saved and completed without an internet connection, and is available in six languages. We also assume no more than basic technical knowledge.

Overview:

Our theory of change is that screening and training supervisors for certain skills that we have found to predict supervisory productivity will result in an increase in women supervisors. This will be due to the fact that there will be a reduction in pre-screening (selection by hiring managers into supervisory pipelines) thereby creating a less biased hiring framework.

It will also have a positive impact on the firm’s business outcomes (such as line-level productivity, worker retention, attendance) and worker’s wellbeing outcomes such as workplace satisfaction and safety. This is based on three pilots (of a total of 300 people) of the entire framework and content, another upcoming pilot involving about 1200 people, a previous full scale randomized controlled trial with about 3000 people and a decade of line-day level administrative data.

1. Input

   1. Psychometric Screening Test

   2. Soft Skill Training

   3. Technical Capability Test

2. Output

   1. New hires/candidates for promotion take screening and technical tests, and receive training

3. Intermediate Outcomes

   1. Reduction in pre-screening of male candidates for supervisory positions

   2. Better managerial composition and know how

4. Final Outcomes

   1. Better overall gender balance at supervisory level

   2. Increased line productivity and workplace satisfaction

   3. Decreased absenteeism and attrition

There are over 45 million people whose livelihoods depend on the textile manufacturing sector and a large proportion of these people are low-income women. Ultimately, over the course of the next five years or more, we seek to positively impact the lives of all these women and improve the work environment for all associated workers. The textile industry is the second largest employer in India after agriculture and any impacts that we can make in this sector will ripple out into the rest of the economy.

Our decade-long partnership with Shahi Exports Pvt. Ltd, India’s largest apparel manufacturer, ensures scale-up opportunities for our project. They have invested in and supported our prior foundational work and have committed to scale up this tool in all their factories post the ongoing piloting process. We are also in conversations with some of Shahi’s large-scale multi-national buyers like H&M and GAP who are eagerly awaiting the evaluation results of the tool to further the scale-up opportunities in their other supply factories. In parallel, Walmart has also shown keen interest in testing our program to see its impacts on the gender gap at the managerial level.

Furthermore, if the proposed study demonstrates significant impacts, we aim to scale this intervention to other sectors with similar workplace gender issues over the course of the next five years or more. This could potentially spread the impact of our technology to the rest of the low-cost mass-manufacturing sector in India and the rest of the developing world.

Our impact goals are ambitious and more long-term, however, in the short term we monitor and measure the impacts of our solution being implemented in the factories of our industry partner through two means: administrative data and field data.

a. Administrative Data: Our partnership with Shahi Exports ensures that we have access to granular administrative data that allows us to track the precise impact that our solution has on worker productivity, attrition rates, rates of absenteeism and gender composition at the supervisory level. This helps us measure the impacts of access to the Pratibha tool on firm level outcomes as well as changes in the gender composition of the managerial cadre.

b. Field Data: Our surveys and interviews will help us measure wellbeing impacts on workers who are managed by supervisors who have had access to the Pratibha tool. We’ll use survey instruments to gauge the workers’ workplace satisfaction and the supervisors ability to resolve conflict, plan work, encourage workers, etc.

In addition to these measures we will also be conducting endline surveys of the workers who are employed in the factory at that time once every two years for the next 6 years.

1. Financial - Though our prototype has been developed and piloted under limited conditions we are still rolling out our large-scale trial in 50+ factories of our industry partner. However, such a large trial requires a great deal of capital investment to buy the equipment (like tablets) that we need to successfully carry out these proof-of-concept studies. We have been struggling to raise this initial capital and this has slowed down our progress.

2. Technical - we need a technical team to manage the backend of the operation as we scale it up to 3000 supervisors and 12000 workers over 2022. Also, given that the tools are based on tablets, there is likely to be some inertia to administering and taking the test that we may not be able to pick up with learning effects. To better monitor and update our tool we will need to expand our technical team.  

3. Cultural - Since we are making additions to the hiring paradigm, there can be an expectation of resistance from the hiring managers leading to lower adherence.

4. Market - Due to our solution still being in its early stages, we have been unable to attract more firms to subscribe to our tool.

We plan to address the barriers we face in the following manner:

1. Financial: We are approaching various sustainable development initiatives and gender justice funds to obtain the monetary resources required to finish our large-scale proof-of-concept studies. The seed money received from our industry partner, Shahi Exports, has taken us quite far however we are currently in the process of diversifying our portfolio of financial backers so as to make the development process of our platform more sustainable.

2. Technical: This barrier is linked to the financial barriers we are in the process of addressing. To increase the size of our backend technical team, so that it can keep up with the increased pressure on the platform, we will need additional monetary resources. Once they come through our first priority will be to add full-stack developers and engineers to the team.

3. Cultural: We are trying to pre-empt any resistance from the existing hiring and training paradigm within these factories by consulting with elements of the pre-existing structure when developing, piloting and finessing our technology. By involving them as stakeholders in this process we seek to overcome any hesitancy they may have with reference to the Pratibha tool.

Most of the following team is part of Good Business Lab (GBL):

Principal Investigators:

1. Achyuta Adhvaryu

2. Anant Nyshadham

3. Jorge Tamayo

4. Jean-Francois Gauthier

5. Emir Murathanoglu

6. Smit Gade

7. Anant Ahuja

Field Research Team:

1. Varun Chati

2. Mohit Verma

3. Bopanna Changappa

4. Karthik

5. Rajneesh Singh

Product Team:

1. Shreepad Patil

2. Mansi Kabra

3. Mamta Pimoli

4. Pranab Prakash

The Pratibha tool has been designed, developed and tested by a team of academics, industry specialists and development sector professionals who have decades of experience in labor research across the developing world. Their extensive business, research and design portfolios ensure that they are best placed to develop a tool designed to help women in the Indian textile industry. The team has worked with garment workers for over a decade and has carried out important studies that have informed the development of this solution. They have evaluated the training component of the Pratibha tool via a large-scale RCT across 25 garment factories where they found large and persistent impacts as a result. Next, they also led an implementation process pilot of the Screening tool among 60 supervisors across 3 factories in order to better understand problems in the tool’s interface and to gauge its user friendliness.

Finally, one of the co-founders of the team is also the head of organizational development at one of India’s largest garment manufacturing firms, Shahi Exports. This helps us establish buy-in for the technology with a key component of the garment industry, something that is essential for the successful take-up and spread of our solution.

GBL is an equal opportunity employer. We celebrate diversity and are committed to creating an inclusive environment for all employees. 

Our Code of Conduct prohibits any discrimination based on race, color, national origin, religion, sexual orientation, gender, gender identity, age, or physical disability. We have zero tolerance for sexual harassment/ exploitation /abuse/misconduct. Our Ethics and Internal Complaints Committees (consisting of an external lawyer) look into such complaints. 

The current leadership team of ten consists of our founders and vertical heads, four of whom are women. This year we piloted 'Anonymized Hiring' and Affirmative Action for several roles to hire candidates from underrepresented backgrounds, and intend to move forward with our learnings. We are in the process of hiring a Diversity and Inclusion Consultant to streamline our strategy. 

Last, our employee well-being focused policies such as flexible leave & work, and individual budgets for expenses such as therapy cement a culture of care and inclusion of different employees’ needs. We encourage employees to proactively share any grievances (anonymous channels available) and by promoting a culture where overwork is not celebrated we ensure that women (or any other disadvantaged social groups) who have competing unpaid work demands on their time can thrive.

Currently we are partnered with Shahi Exports Ltd. Shahi is India’s largest manufacturer of readymade apparels and operates 70 factories across India that employ over 100,000 workers, 72% of whom are women. The solution team has extensive experience working closely with Shahi. The firm has provided the seed capital for the development and testing of this tool and has also provided access to their factories and workers for piloting purposes. After the conclusion of the last round of pilot studies Shahi has promised to scale-up the tool to all its factories across India.

Our tool has two key stakeholders: the firms and their workers (women workers in particular). We seek to create value for both, and aim to produce positive social and financial incentives for the up-take and use of our platform.

For Workers:

Pratibha will help more low-skilled women workers move from low-wage frontline positions to higher paid supervisory roles where they will be able to better address the many issues that plague female frontline workers and help improve the work environment for all workers.

For Businesses:

Creating a safer and more responsive workplace for the majority female workforce will help boost firm productivity while also helping to curb attrition rates. This helps the firm's bottom line since it reduces production disturbances in addition to recruitment and training costs. Our solution hence promotes cash flow efficiency as well as gives firms access to the best candidates for supervisory roles irrespective of gender.

We believe that the challenge can help us overcome primarily market and finances related barriers that have slowed the growth trajectory that we had foreseen for this tool. If we are to win the challenge we are looking forward to having access MIT Solve’s network of resource partners, mentors and coaches who will be able to best advise the solution team on how to market the technology and successfully pitch it to other players in the Indian mass-manufacturing sector. We also look forward to MIT Solve’s advice and guidance on how we can further develop and refine our business model. 

Though our solution team is confident about its research and design work, we could greatly benefit from advice, mentorship, guidance and networking in the business realm so that we will be in a better position to scale-up in the near future. If we are able to market and brand our solution better we believe that our solution would be more successful in attracting external funders and grant makers, thereby helping us overcome our financial barriers too.

Our research team has established that a market space exists in mass-manufacturing contexts for a screening and training tool like Pratibha, however to successfully seize the space we need to develop a sustainable business model and a mature marketing strategy that can attract a wider client base. A tool like ours requires industry buy-in in order to grow and that will not be possible without more attention being focused on these areas. Hence our main goal in partnering with MIT Solve is to attend the 6 month long support program where we seek to gain access to Solve’s network of mentors and partners from various industries. With the right guidance and advice from industry experts we believe that we will be able to refine our business model and strengthen our scale-up plans.