LARC is a research-focused data set that contains information on student demographics. Data in LARC are comprised of 3 categories: 
- Student Info: Data which is constant throughout a student’s academic career (such as date of birth, ethnicity, SAT test scores, high school GPA, etc)
- Student Term Info: Data that can change from term to term (such as term GPA, enrolled credits, major, etc)
- Student Class Info: Data that can change from class to class (such as the total number of students enrolled in a particular class, the grade given by instructors to students, meeting locations, etc)


Learning Analytics Data Architecture (LARC)

Certain student demographic variables can be used to predict future student performance. Rather, it is possible to use a student's current demographic to predict how will they will perform in the future.

Predictive Factors in Academic Achievement

Originally conducted in 1999, the NLSF is a survey issued to several universities across the United States, but centered around the University of California at Berkley. A total of 3,550 students were surveyed in five "waves", and purposely oversampled minorities in an effort to analyze minority underachievement.

National Longitudinal Survey of Freshmen

Cheriyan et. al performed a meta-analysis of over 300 articles to investigate the gender disparity in STEM enrollment. Rather than treating college-level STEM courses as a homogenous whole, Cheriyan et. al investigated differences between individual STEM fields, using a sociocultural perspective. Three key factors that, when combined, contributed to the gender disparity:
- Masculine culture in a field: a social and structural environment that confers a greater sense of belonging and ability to succeed to men than women.
- Lack of early STEM experience: the lack of courses offered or exposure to field in pre-college education.
- Gender gaps in self-efficacy: female students' impression that they are "less competent" than their male counterparts, despite performance metrics that would indicate otherwise.

Gender Disparities in STEM Enrollment

Crues et.al found that the motivations behind a student's enrollment in an introductory CS MOOC correlates with their continued persistence and participation. In addition, the motivations for enrolling in a CS MOOC differ between genders. Data was gathered from a voluntary beginning-of-course survey (n=5807), and analysis was done with k-means clustering and text/data-mining. Notable results were:
- Interaction with the MOOC forums correlated with persistence for both genders.
- Men were less likely to post in the forums and participate in the MOOC when compared with women.
- Men persisted longer in the course.
- Women who were in the computer-science-student cluster were most likely to participate on the forums.

Gender Differences in Learning Goals and Persistence in a CS MOOC

San Jose State University

- Feedback in education
- Achievement gap in education 
- Self regulation in education
- Student demographics
- Academic performance
- Data architecture

Focuses of Learning Analytics

Multiple studies have looked at different aspects of feedback to students: 
- Valence
- Assignment

Learn Before

Related

Learn After