Xiaomin Qiu stands among colorful sculptures of stick figures

Connections: Data and real life

Dr. Xiaomin Qiu remembers sitting in a Chinese classroom during a sixth grade geography lesson. Pointing to a map of her homeland, the teacher explained that 96% of the population of China lived along the east coast.

Qiu, associate dean of the College of Natural and Applied Sciences and a geospatial sciences faculty member, asked a logical question: Why don’t people spread out more evenly?

She imagined streets with less congestion. It felt like the perfect solution.

But it’s never quite that simple, she learned. You have to consider infrastructure, landscape, natural resources and other factors. Variables like these can influence how people interact with their environment and where they choose to live.

“Two scientists can look at the same set of data and have different insights.”

Now Qiu gathers and analyzes data to better understand the connections between variables and outcomes. Recently, this work resulted in several publications about determining potential sinkhole sites based on the collection of new elevation data.

“We’re trying to understand this world in a fully contextualized way, based on all the data and the methods,” she said.

In addition to studies on hazard vulnerabilities, she’s also interested in land use, health and population estimation.

What thread connects her varied research interests? They each have a personal and societal impact.

Predicting MAP success

As a mother with an interest in math, Qiu proposed another research question: What was the greatest contributing factor to successful math scores on the Measures of Academic Progress, or MAP, test throughout Missouri?

You might wonder how this relates to geospatial sciences. She explains that the first law of geography says everything is related. Things that are near to each other are more closely related.

Since students within a school district are physically close together, they are affected by many of the same variables.

“Seeing trends can help us more efficiently use our resources,” Qiu said.

After obtaining scores from 516 public school districts, she identified variables that she suspected might play a role in MAP test scores. She surveyed teacher credentials and explored census data.

Xiaomin Qiu sits in front of her computer studying data

The work of Dr. Xiaomin Qiu has practical applications. She answers questions that can inform policy. Her research can also influence resource allocation – to make the biggest difference to the communities that need it the most.

Would teachers who held master’s degrees be correlated with the highest achieving students? If a teacher had a math teacher certification, would the students reach peak performance?

Or did the family demographics have a greater influence? Would average household income play a role? How would the frequency of single parent households play into the score?

What’s the biggest factor?

Unsurprisingly, previous related studies showed a student’s attendance, participation and homework completion correlated with better test scores.

“Success is strongly determined by personal characteristics, like educational values, confidence and participation in extracurricular activities,” Qiu said.

But analyzing the data made one thing clear: School districts with the greatest household income had the greatest scores.

“When the area has more wealth, public schools receive more taxes, making them a more attractive school to land quality teachers,” Qiu said.

Due to the relationship between higher pay and higher levels of education themselves, Qiu says parents in these areas are more likely to pay closer attention to their children’s academic performance.

“Right now we have so much data. We can take full advantage of the data and produce some meaningful results for people to use. I think this is very important in this field.”

Across the region, though, the significance of individual variables changed.

In the southwest Missouri area, for example, the greatest relationship to student proficiency was parental education level.

In contrast, the greatest influence on math scores for northwest Missouri students was the number of parents in the household, with two-parent households being associated with higher scores.

This proved that one solution wouldn’t improve scores statewide.

“In the end, this study can give policy makers and administrators a new way to think about what they need in different schools,” she said. “We know income, or other influencing factors, are going to be at play, but how big is the role right here? This way you can customize a solution – one that is effective and efficient for your needs.”


One Response

Leave a Reply