S. P.

Mandali’s
Ramnarain Ruia Autonomous College
Matunga, Mumbai - 400019
Department Of Computer Science & IT

Research Project Report

On

Gender Discrimination at Workplace

Course Code: RPSRPCSO605

Guide Name
Mrs. Rasika Mundhe

Project By
Devika Suresh Nair
Seat No - 515

M.Sc. Sem III Computer Science

2024-2025
 S. P. Mandali’s
Ramnarain Ruia Autonomous College
Matunga, Mumbai - 400019
Department Of Computer Science & IT

Research Project Report

On

Gender Discrimination at Workplace

Course Code: RPSRPCSO605

Guide Name
Mrs. Rasika Mundhe

Project By
Devika Suresh Nair
Seat No - 515

M.Sc. Sem III Computer Science

2024-2025
 Acknowledgement

I would like to express my deepest gratitude to everyone who contributed to the

completion of this project.
First and foremost, I am sincerely grateful to my supervisor/ mentor, Rasika Mundhe
ma’am, for their continuous guidance, support, and valuable insights throughout the
course of this research. Their expertise and encouragement have been instrumental in
shaping the direction of this project.
I would also like to thank Ramnarain Ruia Autonomous college for providing the
resources and platform to carry out this study.
I extend my heartfelt appreciation to all the participants who took the time to respond
to the survey and share their experiences. Their candid feedback was invaluable in the
exploration of workplace gender discrimination and the various factors that contribute
to it.
Additionally, I would like to acknowledge my peers and friends for their moral support
and constructive feedback. Special thanks to my family for their unwavering
encouragement and understanding during this process.
Lastly, I express my gratitude to all those who have directly or indirectly helped in the
successful completion of this project. Your support has made this journey a rewarding
experience.

Thank you all.

 INDEX

SR. PARTICULARS Page no.

NO.
1
Abstract 4

2
Introduction 5

3
Literature Review 5

4
Research Methodology 6

5
Experimental setup 10

6
Results 11

7
Conclusion 25

8
References 26
Title: To study the impact of gender discrimination at workplace

Abstract:
Despite significant advancements in gender equality, women remain underrepresented in
executive positions. Therefore, this research will explore gender dynamics in the
workplace using data science techniques.
The aim is to analyse a variety of factors such as promotions, job satisfaction, salary paid
and performance evaluations in order to uncover gender differences and biases. This
research will identify the causes of this imbalance, including societal norms and
organizational cultures in hiring and promotion processes. Using machine learning
models, it predicts career progression and identify patterns of inequality. This study not
only highlights existing issues but also offers data-driven insights to stand-in a more
equitable work environment.
This study aims to understand the effects of gender bias on employees’ career
trajectories. Utilizing a mixed-methods approach, the research collects data through
google forms and employs advanced analytical techniques, including predictive
modelling with Decision Tree, Random Forest, Naive Bayes, SVM and Logistic
Regression. The findings highlight the various forms of gender bias encountered in
professional settings, revealing significant disparities in opportunities and advancement
based on gender. This research underscores the necessity for organizations to implement
effective strategies to combat discrimination, enhance gender diversity, and cultivate an
inclusive workplace culture.

Problem Statement:
The problem statement of gender discrimination at workplace remains a significant
barrier to achieving equality, with women often facing challenges such as unequal pay,
limited leadership opportunities, and biased promotion practices. Despite ongoing efforts
to close the gender gap, many traditional studies rely on qualitative data, which may be
subjective and limited in scope. This project aims to address this issue by using advanced
data science techniques to objectively analyse gender dynamics, identify key factors
contributing to workplace inequities, and provide actionable insights to promote gender
equality in professional settings.
Introduction:
Gender disparities in leadership roles are a significant issue across many industries,
including the tech sector. Despite significant advancements in gender equality over the
past few decades, women and non-binary individuals often face systemic barriers that
affects their professional growth. These barriers manifest in various forms, including
unequal pay, limited access to leadership roles, and discriminatory practices in hiring and
promotion processes. Moreover, gender bias can create a hostile work environment,
leading to decreased job satisfaction and increased turnover rates.
The research uses data science to analyse how gender impacts various aspects of
professional life, including salary, promotions, and job satisfaction. Machine learning
techniques will be implemented to identify patterns and predict outcomes. The findings
of this study will provide insights and recommendations for creating more inclusive and
equitable workplaces. By collecting and processing data from a diverse sample of
individuals, the study seeks to uncover patterns of discrimination and provide insights
into the experiences of those affected. Additionally, the research will highlight effective
strategies for organizations to combat gender bias, promote inclusivity, and foster a
culture of equality.
Ultimately, this study contributes to the ongoing discourse on gender equality in the
workplace, offering valuable recommendations for businesses and policymakers to create
more equitable environments for all employees.

Literature review:
Research indicates that women face multiple barriers in their career progression,
including societal expectations, organizational biases, and a lack of support systems.
Studies have shown that women often earn less than men for the same job, face higher
barriers to promotion, and report lower job satisfaction. For example, a study by Blau and
Kahn (2017) found that the gender pay gap persists even after accounting for factors such
as education and experience. Additionally, research by Catalyst (2019) highlights that
women are underrepresented in leadership positions, a phenomenon often referred to as
the "glass ceiling."
However, traditional studies mainly rely on surveys and qualitative data, which can be
limited in scope and subject to biases. This research aims to fill this gap by using
advanced data science techniques to analyse datasets which allows for more complete and
objective insights into gender dynamics in the workplace. It reveals a complex interplay
of systemic biases and cultural norms that continue to affect employees across various
sectors. Research also indicates that women are underrepresented in leadership positions,
often facing barriers to advancement due to both overt and covert forms of discrimination
(Catalyst, 2023).

Additionally, qualitative studies have explored the experiences of individuals who

encounter gender bias, revealing patterns of microaggressions, stereotypes, and unequal
access to opportunities (Davis et al., 2022).
Turesky and Warner (2020) emphasize the importance of women in management roles,
highlighting how their unique perspectives contribute to more inclusive planning
processes, while Kalev and Deutsch (2018) explore how workplace organizations
contribute to gender inequalities, proposing a focus on institutional reforms to address
these disparities. Marquez et al. (2020) provide a global perspective on gender diversity
in the planning profession, noting significant regional disparities in gender representation.
Alan et al. (2020) explores gender differences in leadership, particularly in decision-
making during adolescence, revealing a decline in girls' leadership willingness as they
age. The comprehensive study by Askari et al. (2024) delves into gender politics at
workplaces, identifying persistent challenges like the gender pay gap and biased
promotion practices. Stewart-Williams and Halsey (2021) discuss gender differences in
STEM fields, advocating for policies that consider both biological and sociocultural
factors. Hebl and King (2019) provide a broad overview of gender bias, emphasizing the
intersectionality of gender with other identities, such as race and socioeconomic status.
Lastly, Oppong and Bannor (2022) conduct a systematic review on gender and power
relationships in Africa and Asia, highlighting the influence of societal and organizational
factors on workplace power dynamics. Together, these studies underscore the ongoing
challenges in achieving gender equality and offer a range of future research directions
and interventions to promote inclusivity and diversity in the workplace.

Research Methodology:
(A)Description of Algorithms:
In this research, several machine learning algorithms were employed to analyse and
predict factors related to gender discrimination in the workplace. Naive Bayes, a
probabilistic model, was used for efficient classification of workplace experiences.
Logistic Regression helped predict the likelihood of gender-based disparities using
multiple input features. Support Vector Machine (SVM) was applied to capture
complex patterns and relationships in the data. Decision Tree models provided
interpretable insights by mapping decision paths that lead to certain outcomes.
Finally, Random Forest, an ensemble method, enhanced prediction accuracy by
aggregating the results of multiple decision trees, ensuring robust and reliable outcomes.

(B) Procedures, Data Collection Methods:

The data for this study was collected through a Google Forms survey, designed to
capture both quantitative and qualitative insights from respondents regarding their
experiences and perceptions of gender discrimination in the workplace. The respondents
were required to answer both multiple choice questions and personal response
questions. Nearly 150 responses were gathered and dataset was then processed to handle
missing values, duplicates and other inconsistencies and were converted as required.
Google form questions: Gender Discrimination in Workplace (google.com)
(C) Logic, and Process Flow:
The logic and process flow of the project began with data collection from a survey of
workplace experiences related to gender discrimination. The data was then cleaned and
pre-processed, addressing missing values and outliers. Following this, exploratory data
analysis (EDA) was conducted to identify patterns and trends in key areas such as work-
life balance and leadership opportunities.
For the prediction phase, various machine learning models such as Naive Bayes, Logistic
Regression, SVM, Decision Tree, and Random Forest were applied using a 70/30
train-test split. The models were trained on 70% of the data and tested on the remaining
30% to predict outcomes like workplace satisfaction and discrimination. Additionally,
sentiment analysis of open-ended responses provided deeper qualitative insights. The
entire process was integrated into an interactive user interface (UI) built using Streamlit,
allowing users to visualize the findings and make predictions based on real-world data.

Experimental setup:
Tools, Architecture/Framework and Software Language:
The research project employs a range of tools and technologies to effectively analyse
gender discrimination in the workplace. Python is used as the core programming
language, given its extensive availability of libraries and ease of integration. Data
collection is facilitated using Google Forms for survey design, while Excel aids in
preliminary data organization.

For data manipulation and preprocessing, libraries such as Pandas and NumPy were
utilized to clean, organize, and transform the raw data collected via Google Forms.
Scikit-learn library provides machine learning algorithms, such as Random Forest, Naive
Bayes, and Logistic Regression, for predictive modelling. Data visualization is
accomplished using Matplotlib and Seaborn packages, which help create clear and
informative graphical representations of the findings. The architecture of the project
includes a Data Collection Module for survey design and response storage, a Data
Processing and Analysis Module for cleaning and modelling data, and a Visualization
Module for generating insights. Sentiment analysis on open-ended questions was carried
out using the Natural Language Toolkit (NLTK) to extract insights from textual
responses, identifying underlying emotions and patterns. An interactive user interface is
built with Streamlit. It was chosen for its ability to integrate Python scripts with the front-
end, allowing users to interactively upload datasets, visualize results, and view model
predictions in real-time. It enables the users to explore the data and results seamlessly.
Throughout the project, Python serves as the primary programming language, using its
extensive libraries and frameworks to facilitate data analysis and visualization effectively.
The project’s architecture follows a client-server model, with the UI developed using
Streamlit. The entire project is hosted locally, offering a streamlined, interactive platform
for exploring and analysing workplace gender discrimination data.

Results:
UI screen layout:
Home page-

Upload data page-

Eda page-

Summary statistics:
Demographic analysis:
Workplace experience analysis:
Work-life balance analysis:
Leadership and gender analysis:

Comparative analysis:
Predictive modelling page-

ML model: Target variable for prediction:

Sentiment Analysis page-
Prediction (testing) report:
Results:
Here the dataset is divided into two parts, with 70% allocated for training the model and
30% randomly selected for testing. The testing data is then used to evaluate the model's
performance by predicting the target variable, where the predictions are compared against
actual outcomes to assess accuracy based on the other influencing factors. The analysis
revealed key insights into the prevalence and impact of gender discrimination in the
workplace. The results observed is that the accuracy of Random Forest is the highest for
all three target variables.
When the target variable is– ‘have you ever experienced or witnessed gender
discrimination at work’, the model with highest accuracy rate is Random Forest with
the accuracy of 0.76.
When the target variable is– ‘how satisfied are you with your work-life balance’, the
model with highest accuracy rate is Random Forest with the accuracy of 0.56.
When the target variable is– ‘do you feel that your workplace provides equal
opportunities for advancement regardless of gender’, the model with highest accuracy
rate is Random Forest with the accuracy of 0.67.
Finally, sentiment analysis of open-ended responses further highlights the emotional toll
of discrimination, with negative, positive and neutral sentiments. For better
understanding, the sentiment analysis is done based on the age group, respondent’s
gender, their years of experience and the industry they work in.

Conclusion:
Findings of project work:
The findings from this project reveal significant trends and patterns related to gender
discrimination in the workplace. The data analysis showed that a considerable portion of
the respondents reported facing unequal opportunities for advancement, with women
being disproportionately affected compared to their male counterparts. Work-life balance
satisfaction was also a key issue, particularly for those experiencing gender-based
challenges in leadership roles and workplace dynamics.
Through the predictive models, we were able to identify key factors influencing these
experiences, such as job roles, leadership representation, and the perception of
organizational support. The machine learning models used in this study, including Naive
Bayes, Logistic Regression, SVM, Decision Tree, and Random Forest, successfully
predicted outcomes related to discrimination and satisfaction, offering valuable insights
for addressing these disparities.
Overall, the project underscores the importance of fostering an equitable work
environment and highlights areas where organizations can focus their efforts to reduce
gender-based discrimination. Future work could expand on these findings by
incorporating a larger, more diverse dataset and exploring additional factors that
influence workplace equity.
Future enhancements:
There are several opportunities for expanding and improving this project in the future.
● Increasing the dataset size by gathering more responses from diverse industries
and geographical locations, which would provide a more comprehensive
understanding of gender discrimination across different sectors.
● Incorporating more advanced machine learning algorithms, such as deep learning
models, could improve the predictive accuracy of the outcomes.
● Exploring other variables, such as race, education level, and company size, could
offer deeper insights into the intersectionality of workplace discrimination.
● Integrating real-time data collection and analysis could help organizations actively
monitor and address issues of gender discrimination as they arise, ensuring a more
dynamic and responsive approach to workplace equity.

References:
Research papers:
Gender Dynamics in the Planning Workplace: The Importance of Women in
Management by Marisa Turesky and Mildred E. Warner in March 2020 acquired from
Research gate-
https://www.researchgate.net/publication/
339835705_Gender_Dynamics_in_the_Planning_Workplace_The_Importance_of_Wom
en_in_Management

Gender Inequality and Workplace Organizations: Understanding Reproduction and

Change by Alexandra Kalev and Gal Deutsch in June 2018 acquired from Research gate-
https://www.researchgate.net/publication/
325605384_Gender_Inequality_and_Workplace_Organizations_Understanding_Reprodu
ction_and_Change
Gender Diversity: An opportunity for socially inclusive human resource management
policies for organizational by Caterina Galdiero acquired from MDPI
Gender Diversity: An Opportunity for Socially Inclusive Human Resource Management
Policies for Organizational Sustainability (mdpi.com)

Gender Diversity in Leadership: A Bibliometric Analysis and Future Research Directions

by Elisabete Vieira, Mara Madaleno, and Júlio Lobão in the year 2020 acquired from
MDPI
https://www.researchgate.net/publication/
361965297_Gender_Diversity_in_Leadership_A_Bibliometric_Analysis_and_Future_Re
search_Directions

Understanding Gender Differences in Leadership by Sule Alan, Seda Ertac, Elif Kubilay,
Gyongyi Loranth in the year 2020 acquired from Oxford Academy.
Understanding Gender Differences in Leadership | The Economic Journal | Oxford
Academic (oup.com)

Gender Politics at Workplaces: An In-Depth Exploration of Challenges, Strategies, and

Societal Implications by Syed Shuja Askari, Dr. Shobhana Singh, Sanjeev Kumar Pandey
in the year 2024 acquired from jetir.org.
JETIR2402532.pdf

Men, Women and STEM: Why the Differences and What Should Be Done? by Steve
Stewart-Williams and Lewis G. Halsey in January 2021 acquired from Sage Journals.
https://www.researchgate.net/publication/
348295173_Men_Women_and_STEM_Why_the_Differences_and_What_Should_Be_D
one

Gender Equality in the Workplace: An Introduction by Mikki Hebl and Eden B. King in
the year 2019.
https://www.researchgate.net/publication/
337520376_Gender_Equality_in_the_Workplace_An_Introduction

Gender and Power Work Relationships: A Systematic Review on the Evidence from
Africa and Asia by Dixon Oppong & Richard Kwasi Bannor in the year 2022.
Gender and power work relationships;: A systematic review on the evidence from Africa
and Asia (tandfonline.com)

Websites:

5 Ways to Improve Gender Equality in The Workplace | World Economic Forum

(weforum.org)
Gender inequities in the workplace: A holistic review of organizational processes and
practices - ScienceDirect
Gender inequalities in the workplace: the effects of organizational structures, processes,
practices, and decision makers’ sexism - PMC (nih.gov)
Women in the Workplace 2024 report | McKinsey
Why Gender Equity in the Workplace is Good for Business - Professional & Executive
Development | Harvard DCE