A Cautious Statistical Look at Using Family Factors to Predict Educational Outcomes

• In Event: AERA Roundtable Session 15
  In Roundtable Session: Applications, Demonstrations, and Evaluations of Statistical Models and Methods

Abstract

Running three statistical models on a dataset, I found out that among the components of SES (parental education, parental occupation, and household income), only education significantly contributes the most to SES. Education significantly predicts the odds of parental involvement in students’ education. For parental involvement, I employed attending school conference as a proxy and found out that it significantly predicts students’ performance in math and reading. The findings suggested that there might be misuses of SES components and an ignoring of parental involvement in the relationship between SES, especially parental education, and students’ educational outcomes. Therefore, studies aiming to predict students’ educational outcomes should handle family factors with care.

Author

• Binh Chi Bui, University of Houston