[go: nahoru, domu]

Skip to content

UBC-MDS/DSCI_532_group26

Repository files navigation

An App to report Survey on Mental Health in the Tech Workplace

Contributors Github
Bruhat Musunuru @bruhatM
Siqi Zhou @roycezhou
Chun Chieh(Jason) Chang @jachang0628

Dashboard link

https://mental-health-dashboard-ubc.herokuapp.com

About

Our app aims to present the statistics of people who worked in the tech sector and their mental illness history versus employers' attitudes towards mental illness across different states in the USA. We hope that by presenting these statistics, employers around the world can stop overlooking the issue of mental illness and create a less stress-inducing work environment.

Description

This APP has a dashboard that summarizes the information given by the participants who took the mental health survey. In the right panel, there are multiple tabs that contain different graphs. In the Mental Illness History tab, we have plots to represent whether an individual's mental illness interferes with work, how often that an individual talk to his/her supervisors, and whether an individual seeks treatment for mental illness or not. In the Company Culture tab, we have plots to represent whether the employer takes mental illness seriously, the easiness to leave work due to mental reason, and the workplace benefits between states. In addition, we can further filter the above graphs using state, gender, and the ability to work remotely. Finally, the third tab has the demographic information regarding all the participants. Note that the graphs in the third tab are static so they won't be affected by the global widget.

In the left panel, the users can filter the states they are interested in. All filters will directly affect the 6 plots produced on the right. By choosing a state, the plot will only show the data that is restricted to that particular state. By selecting the gender, all 6 plots will only include the specific gender (male, female, all) with the corresponding mental health data. Hence, users can fix a certain gender that they are extremely interested in. Similarly, users can use the remote work button to filter data for all 6 plots as well. They can also choose to show or hide the bars of the plots by clicking on the corresponding legends.

With these information, we hope to answer the research questions such as how do the number of people who sought treatment compare between states, how easy is it to leave work due to mental health issue, and how does the frequency of mental illness that interfered with work vary across states etc.

Initial app sketch

Alt text

Dashboard in action

Alt text Alt text Alt text

Usage

Access dashboard at https://mental-health-dashboard-ubc.herokuapp.com

To build a local dashboard

Step 1:

Clone this repository to your local PC.

Step 2:

Create a conda environment using environment.yaml at the root of this project.

conda env create --file environment.yaml

Step 3:

Activate a conda environment by running the following command at the root directory of the project.

conda activate dash_env

Step 4:

Go to the root folder of the repo and execute python src/app.py.

Reference

Mental Health in Tech Survey dataset