2014 - Breaking it Down: Engineering Student STEM Confidence at the Intersection of Race/Ethnicity and Gender

Attribution: Litzler, Elizabeth, Samuelson, Cate C., & Lorah, Julie A.
Researchers: Cate C. SamuelsonElizabeth LitzlerJulie A. Lorah
University Affiliation: University of Washington
Email: elitzler@uw.edu
Research Question:
This study examines social cognitive influences on engineering student STEM confidence, with a particular focus on women and underrepresented minorities.
Published: Yes
Journal Name or Institutional Affiliation: Research in Higher Education
Journal Entry: Vol. 55, No. 8, Pp. 810-832
Year: 2014
Findings:
  1. Although some underrepresented groups may have lower STEM confidence overall, this finding no longer applies to all groups after controlling for personal, environmental, and behavioral factors. Specifically, African-American and Hispanic men report higher average STEM confidence than White men after controlling for these associated measures.
  2. White women continue to report lower average STEM confidence than White men after controlling for these measures, while other groups do not differ from White men.
  3. Both Hispanic and African-American women report similar levels of confidence to White men after controlling for these measures.
  4. Given the disproportionate academic, economic, and social disadvantages these students often face in high school, overcoming these challenges and being selected to enroll in engineering could indicate that they already had high levels of STEM confidence.
  5. Many elements of student perception, including student views of professors, comparisons to peers, perceptions of the field as rewarding, and desirability of chosen major are positively associated with student STEM confidence.
  6. The changing patterns of significance for race/ethnicity and gender groups between the two models indicate that personal, environmental, and behavioral factors have different relationships with STEM confidence levels for different groups.
Scholarship Types: Journal Article Reporting Empirical ResearchKeywords: EngineeringGenderIntersectionalityRaceSelf-ConfidenceSelf-EfficacySTEMRegions: NationalMethodologies: QuantitativeResearch Designs: Secondary Survey DataAnalysis Methods: Descriptive StatisticsMultilevel Models Sampling Frame:Undergraduate Engineering Students
Sampling Types: Non-Random - PurposiveAnalysis Units: CollegeStudentData Types: Quantitative-Longitudinal
Data Description:
  • Social cognitive theory emphasizes how human functioning is shaped by (a) personal attributes, such as internal cognitive and affective states, (b) environmental events, and (c) behavior, all interacting to determine motivation and actions.
  • The data for this research came from a large-scale 2008 online survey of 10,366 undergraduate engineering students from 21 colleges. The majority of the 21 engineering schools are public schools (81 %), although some small and private schools are included. 7,833 students were included in the analysis after list wise deletion.
  • The main predictor of interest is race/gender and includes 12 distinct groups. White men were the reference category, and each intersection of race/ethnicity and gender was included as a dichotomous variable.
  • Six predictor variables represent the three areas hypothesized by social cognitive theory to have a relationship with confidence. Environmental factors which include having good professors, student community, and comparison to peers; Personal factors are is engineering rewarding and major desirability; the behavioral factor is cumulative self-reported GPA. Of the six measures, four were created from multiple survey items, while comparison to peers and GPA were single survey items. All survey items used to create these aggregates were measured with response categories based on a five-point Likert scale.
  • The outcome of interest was student STEM confidence. Three survey items were averaged and then standardized to create the STEM confidence variable. The three items asked about confidence in a student’s ability to succeed in college engineering courses; college science courses; and college math courses. Students responded to each item on a five-point rating scale.
  • The following demographic variables were included as control variables: class year (1st year, 2nd year, 3rd year, 4th year, 5th year, 6th year and beyond); whether the student was a first-generation American citizen or not; whether the student used need-based scholarships to pay for college; and the student’s major field of study.
Theoretical Framework:
Relevance:Factors Related to STEM Readiness
Archives: K-16 STEM Abstracts