Education is our Passport to the Future, for Tomorrow belongs to the People who prepare for it Today!

We all understand the importance of Education and how it plays an important role in shaping our Life. With A Brighter Future, we present our analysis on the U.S. high school graduation rate, which aims to provide key insights on factors affecting the rate as well as actions to improve it.

A Brighter Future also has a web/desktop application, with rich visualizations and dashboards for parents, teachers and the government to understand the factors affecting the graduation rate in their area, state and across the Nation and thereby, analyse it with ease for determined actions.

Using the Insights derived from A Brighter Future, Government, Communities and School Districts can analyse where they should focus more and undertake awareness / support programs in order to improve the graduation rate.


The U.S. national on-time high school graduation rate was only 71.7% in 2001, which went up to 81.8% based on the progress over the last decade. The GradNation campaign has a goal of 90% on-time completion by 2020, for which we need to understand the factors which affect the graduation rates and thereby take actions for a cohesive environment which encourages students to complete their graduation.

And so, with A Brighter Future, using data driven analytics and IT solution, we are aiming to provide insights on the US education system, which can help improve the graduation rate and thereby, prepare the students for a Brighter Tomorrow.


We started with the pre-built dataset provided by the challenge, which includes high school graduation data at the school district level, which was joined with the maximum overlapping Census data. In addition to the merged dataset provided by the challenge, for school districts and their respective graduation rates, we looked in to various other datasets and merged them at the school district level.

Below datasets were analysed:

  1. Graduation Rates
  2. Census 2010 data
  3. American Community Survey 2012 - 5 Yr estimates
  4. U.S. Common Core Data for Local Education Agency
  5. Financial Survey Data for Local Education Agency

We also looked at below datasets:

  1. Juvenile Crime Data
  2. Bullying and School Victimization Survey Data
  3. Overall Crime Data
  4. Child Health Data
  5. Environment Data

However, these datasets were not available at a granular level for a quantifiable co-relation.

Data Model and Visualizations

Following Steps were followed:

  • Data Mashup: Import All Datasets in Relational Database; Create Join Keys and SQL Query to obtain a Single DataSet for Analysis
  • Data Cleansing: Cleansed the Data for Null Values, Missing Values, Ranges converted to Average; Identified 389 Variables for Analysis
  • Pearson Co-relation: Grouped the Variables into related categories; Using Rattle for R, determine the Key Dependent Variables using Pearson Co-Relation. 114 Variables identified
  • Linear Regression: Used Multi-Variable Linear Regression to determine the Model for Predicting the GraduationRate. Adjusted R-squared Value: 0.5938
  • Random Forest Modelling: Used Random Forest Modelling to determine the Predictive Model. % Variation explained: 80.04%
  • Cross Verification: Cross Verification of Models using training datasets and datasets under study
  • Visualizations: Built Visualizations using QlikSense Desktop for data-driven insights, analytics and reports

Features of Data Visualization Application:

  1. Completely Interactive Maps, Bar Charts, Scatter Charts, Gauges for a Rich Visualization Experience
  2. Filters for Grad Rate, Cohort Size, Demographics like State, County, School Districts
  3. Using self-discovery mode, users can play with the visualizations to achieve further insights, compare school districts and so on
  4. Get details of the high school graduation rate by State and School Districts
  5. Identify the characteristics of their respective / prospective Schools Districts
  6. Identify over-achieving and under-achieving Schools Districts
  7. Co-relate data of various identified factors affecting the graduation rate
  8. Government and School Districts can analyse where should they focus more and undertake awareness / support programs in order to improve the graduation rate
  9. Easy to use and update with latest data; free to use on Desktop. Can also be made available as a Web Application

Insights and Suggested Actions

Category Factors Affecting Graduation Rate Insights Suggested Actions
Student Origin, Language Proficiency, Economic background Economically Disadvantaged Students; Students with Disabilities; Students with Low English Proficiency; Black & Native Americans; Average Students / School; White Americans and Asians Economically disadvantaged students, Students with disabilities, students with difficulty in English Proficiency, Black students & native American students are amongst the highest for high school dropouts. Hence, it is imperative to ensure support to such students for a better future. Schools with a Higher Students number had a better graduation Rate. Also, Whites and Asians were amongst the lowest to dropout from high school Government and Schools should support the Education System by: 1. Providing Financial Assistance where required; 2. Improve English Proficiency through Support Programs; 3. Provide a cohesive environment to all strata of students to encourage education and ensure equality to all
Population Demographics Rural Population; Male Population Dominance; Black & Native Americans Population Dominance; Urban and Urban Cluster Population; Female Population Dominance; White Americans and Asians Population Dominance Areas with Higher Rural Population, Male Dominated Areas and Black & Native American Population Clusters have a Higher Drop-out Rate. Areas with Higher Urban Population, Female Dominated Areas and White & Asian Population Clusters have a Lower Drop-out Rate. Females, possibly Mothers play an important role in education of their children. Government and Schools should undertake Awareness Programs in Rural Areas and Areas where the population demographics have lower Graduation Rate. Government should further study the reasons of lower graduation rate, where the Male Population is dominant
Employment Below Poverty Line Population; Already Lower High School Grads Population; Unemployment Rate for Age Group 16-44; Population having College Degree; Employment Rate for Age Group 16-44 Areas where the % population Below Poverty Line, Areas where there were already high number of low-education levels and Higher unemployment added into drop-out rates. This meant that population where already under-priviledged and not employeed or well educated, further fostered lower graduation rates. Population with higher degrees and employment led to better graduation Rates which proves that they encourage education and support the students of their family / community Government, Schools and Communities can undertake programs to: 1. Provide Financial Assistance; 2. Increase Awareness on Benefits of Education; 3. Increase Employment Opportunities
Language Spoken at Home Languages Other than English; Native Citizen of US Families where English is not Native, meant higher drop-outs. Possibly difficulty in getting education in English discouraged the students Schools & Community should undertake special programs to help improve the English proficiency. Government can support such programs economically so that population with lower income can also benefit
Marital Status No Spousal Relationship; Single Parents; Single Mothers; Householders Living Alone or Non-Relatives; Married Couples; At least 2 members are related by birth, marriage, or adoption Households where there were no spousal relationship, or with Single Parents / Mother or Living Alone led to higher drop-outs. This means that Strong Family Bonds are necessary for a cohesive environment for students to pursue education and a better life ahead. Household with Married Couples or those staying with Family Relatives led to higher graduation rate. The community should understand the importance of Family Life and Husband-Wife bonding, which is important for a better upbringing of their kids. Government can also strengthen the support provided to students living with single parent households
Family Dynamics and Income Families receiving Public Assistance Income; Grandparents Responsible for Students; Mean & Median Family Income Households which received more Public Assistance Income, or where Grandparents were responsible for Students led to higher drop-out rates. Possibly Public Assitance Income makes the population lethargic to pursue further education. Higher Income Families had a higher graduation rate which shows that they support and encourage Education Government should review the Public Assistance Income program to ensure that the funds are used in a way that it does not discourage the population from pursing education. Also, Government should support Families where parents are not engaged in the upbringing of the kids, which can also lead to children being sent to orphanage.
Student Mobility Students with Higher Geographical Mobility in Past 1 Year Students who keep on moving from one place to another tend to get discouraged and lead to higher drop-outs. This could be because of difficult process for enrollment to schools Transfer and Enrollment of students from 1 school to another needs to be made easier and streamlined. This can be achieved through a centralized and digitalized system
Teachers and School Staff Average Teachers and Staff / Student More Teachers and Staff does not necessarily mean that the graduation rates will improve. Schools should focus more on the needs to the students which can encourage them to study. New methods of education, digitalized e-learnings can be introduced to encourage students to pursue education
School District Revenues and Expenses School District Revenues / Student; School District Expenses / Student Spending More or Earning More, does not contribute towards a Higher Graduation Rate Schools should look into the areas of Expenditure and analyse where to spend more in order to improve the graduation rates. This can be into areas like After School Programs, Support to schools, Child Counselling Programs and so on

Download the Complete Analysis and Report for an in-depth analysis.

With A Brighter Future, we hope that Government, Schools, Teachers, Parents and Researchers would be able to use the Insights provided and through the IT Application, further derive insights for a Brighter Future for the Students!

Built With

  • mssql
  • qliksense
  • r
  • rattle
Share this project: