ezgif.com-resize (1).gif


P (Education) is a data visualizing tool that addressed non-conventional form factors in educational attainment. It aims to help policy makers to make better decisions by understanding how the long-neglected social-cultural events such as gender and parental income influences educational mobility. This is an independent project based on my five-month ethnographic field research at a rural primary school in Yunnan, Southern China.

Timeframe Mar- July 2018 (Field Research) Sep - Oct 2018 (Design)

Tools D3.js, Tableau (Visualizing Data), Sketch, Principle (Interaction Design)




Source: China National Census 2010

Source: China National Census 2010

According to national census data, more than 50% of eighth-graders in rural China has IQ score of lower than 90. Among all the works from rural China, only 24% attended high school. With China’s booming economy placing new demands on its workers, this becomes one of the biggest challenge that China faces - more than 400 million people in the labor force come from rural areas, they are incapable of learning or/and switching jobs.

In the past thirty years, policy makers and institutions have been striving to improve healthy cognitive development and educational mobility of children from disadvantaged socio-economic backgrounds. Most educational policy instruments derive from research that focused on tangible form-factors such as parental investment in children’s education, bequests from parents and access to parent’s social network. However, a significant part of the inter‐generational correlations presumably arises from the effects of intangible factors, such as genetic transmissions of ability and preference from parents to children, and this still poses challenges in understanding potential policy-making opportunities in this field.


As a sociologist and a designer, I always wanted to leverage my design skills to help tackling social problems. Education is the most concerned topic to me of all time. In early 2018, I reached out to a rural primary school via a NGO and conduced a 5-month field research that aims to explore the dynamics, interactions, potential casual effects, and design opportunities in rural education.

Research Background


The Village

  • Location Yunnan, Southern China

  • No. of Residents 5910

  • Age between 0-16 1082

  • Age between 17 - 55 4016

  • Gender Ratio F:M 1:1.61

  • Avg. Income $3100 / year

  • High School Rate 0-18 5.11%

  • Key Income Resource Agriculture


The School

  • Students : Teacher 167 :5

  • Avg. Size of Class 35

  • Enrollment Rate 97.9%

  • Avg. Test Score 55 out of 100

    (urban: 79 our of 100)

  • Left-behind Children** 92%

I was introduced to local government officials who was responsible for education and was granted permission to conduct field research while teaching at the primary school.

Left-behind children** (留守儿童) refer to children who remain in rural regions while their parents leave to work in urban areas. These children are usually taken care by grandparents or family friends, who remain in the rural regions. As a result of limited access to social and economic infrastructures and inadequate interaction with parents, many of these children face developmental and emotional challenges.


Research Subjects

My key research subjects are students from the primary school. Through day-to-day interactions, I was able to observe and collect first-hand insight of how they study, play, think and live.

Research subjects also include stakeholders such as teachers, the students’ guardian (grandparents), and local government officials who responsible for education.

* Qualitatively research database includes interviews of 35 students of age 7-12, 7 household, 16 teachers and 9 governors

* Qualitatively research database includes interviews of 35 students of age 7-12, 7 household, 16 teachers and 9 governors


Research Methods

1. Participatory Observation

During the day time, I worked as full-time teacher teaching Chinese, English, Arts, Music and PE class. I also lived at school dorm with local teachers and shared life with villagers. During the weekends, the teachers would give me a ride to downtown, and I would join dinner with local government officials. I had an immersive experience and closely observed and documented the daily activities, living conditions and interaction of students, teachers, villagers, and local governors.

2. Semi-structured, in-depth Interview

I conducted semi-structured interviews with students, their guardians, school teachers, and local governors (consensus obtained). For students specifically, the interviews were loosely structured (chatting) as children of young ages might got nervous and uncomfortable of formal discussion. To help engaging conversations, I intentionally organized student group discussion instead of 1:1. The interview questions include but not limited to:

  • With students

    • Do you have siblings/how many siblings?

    • Where are your parents? What do they do for living?

    • What do you want to do when you grow up?

    • How often do you meet with your parents? (if not living with parents)

  • With student guardians

    • How many children do you have?

    • Where are they living now?

    • How many grandchildren do you need to take care of?

    • What is your prospect to your children/grandchildren?

  • With teachers

    • How long have you been teaching here?

    • Where did you go for school?

    • Why do you want to be a teacher here?

    • What do you think is the biggest challenge in rural education? (open-ended question)

  • With governors

    • What do you think is the key differences between urban and rural education?

    • How to you feel about the different dynamics in rural and urban China?

    • Why there is a huge educational achievement gap (according to your view)?

3. Quantitative Data Analysis

As part of the research proposal, I was granted with access to local census and longitudinal educational data. The data dimensions include but not limited to:

  • Enrollment rate (primary school, middle school and high school)

  • No. of resident per household

  • Teacher/student ratio

  • Avg. Score

  • No. of students whose parents work in the urban areas

  • Parental education attainment

  • Parental yearly income

Some of the data has been published yet. Therefore I can’t disclose the details. In the following section, I will conclude the findings as part of the insight which informs my design decision.


Research Findings

Combing qualitative and quantitative research, the findings converged to four directions. they are:

* Intangible factors that are not previously specified in literatures

* Intangible factors that are not previously specified in literatures


In socio-cultural studies, it is commonly acknowledged that there is no “independent” factor. Everything is inherently interconnected. Therefore, to understand the dynamics and potential casual effects, it is important to incorporate every aspects of form-factors to explore how they might impact each other.


Ideation: to helps policy makers to know how to intervene

(and ultimately help to improve education in the rural areas)

From the 5-month research, interviews and conversations with stakeholders, I noticed the biggest challenge that policy-makers face is that they did not have a panoramic data analyzing tool, therefore they were incapable of understanding, communicating or concluding the relationships of different form factors in rural education attainment and intergenerational educational mobility.

To solve this problem, I came across this design idea of incorporating and visualizing related data from different aspects, and give the navigator to the policy maker to explore the potential casual effects. This data visualizing tool aims to present a truthful and pragmatic overview that helps policy makers to make strategic intervention in the complex nature of educational mobility.


I started the design process with low fidelity sketches. This is the way I iterate through many design options quickly.

  • Main purpose of the sketches: brainstorming presenting ideas, communicate with others.

  • I made 3 different versions based on feedbacks and critiques.

  • Along the way of sorting data, I iterated the wireframes by only keeping data that shows the strong correlations (based on research).

  • By using slider bars to navigate different form-factors, policy makers are able to have a holistic view how different factors might intercorrelated.

  • I was advised to input and visualize data of different geographies to give more levels of comparison and insights.

  • The representation of geographic map evolves from plain map to dotted view.


How does the tool work

Group 6.png


Data visualizing tool: D3.js  Prototyping too: Sketch, Principle

Data visualizing tool: D3.js

Prototyping too: Sketch, Principle


This tool has now been translated to Chinese. I am actively communicating with local governors and NGOs to integrate this data visualizing tool to policy-making process.

If you have any ideas, feedbacks, or connections to scale the impact of this project. Please don’t hesitate to reach out!