Skip to main content

Table 1 Descriptive statistics of control variables for the analytic sample with multiple imputation

From: Perceived abilities or academic interests? Longitudinal high school science and mathematics effects on postsecondary STEM outcomes by gender and race

 

Mean

S.E

Min

Max

% Missing

Student characteristics

 Gender

  Male

46.4%

0.0

0.0

100.0

0.2%

  Female

53.6%

0.0

0.0

100.0

0.2%

 Race/ethnicity

  Asian

4.5%

0.0

0.0

100.0

3.8%

  Black

12.2%

0.0

0.0

100.0

3.8%

  Latina/o/x

20.3%

0.0

0.0

100.0

3.8%

  Multiple/Other Race

8.5%

0.0

0.0

100.0

3.8%

  White (Reference)

54.6%

0.0

0.0

100.0

3.8%

 Family characteristics

  Family income

4.61

0.1

1.0

13.0

19.8%

  Father’s occupation (STEM)

15.3%

0.1

0.0

100.0

34.6%

  Mother’s occupation (STEM)

14.5%

0.1

0.0

100.0

26.3%

  Parental education

     

  4-year college degree (Reference)

44.1%

0.0

0.0

100.0

19.7%

  No 4-year college degree

55.9%

0.0

0.0

100.0

19.7%

 Student preparation and performance

  Highest math course taken

1.61

0.0

0.0

2.0

4.4%

  Highest science course taken (Physics pipeline)

0.5

0.0

0.0

2.0

4.4%

  HS math course GPA

2.6

0.0

0.0

4.0

4.7%

  HS science course GPA

2.7

0.0

0.0

4.0

4.9%

High school and college characteristics

 School types

  Public (Reference)

90.3%

0.0

0.0

100.0

0.0%

  Catholic

5.1%

0.0

0.0

100.0

0.0%

  Other private

4.7%

0.0

0.0

100.0

0.0%

 School location

  Urban (Reference Group)

31.9%

0.0

0.0

100.0

0.0%

  Suburban

34.7%

0.0

0.0

100.0

0.0%

  Town

10.1%

0.0

0.0

100.0

0.0%

  Rural

23.2%

0.0

0.0

100.0

0.0%

 Type of college first attended

  Four-year (Reference)

63.0

0.9

0.0

100.0

1.0%

  Two-year

37.0

0.9

0.0

100.0

1.0%

Classroom characteristics

 Experiences with math teacher

3.1

0.0

1.0

4.0

15.7%

 Experiences with science teacher

3.1

0.0

1.0

4.0

20.9%

  1. The analytic sample size is n = 11,560. S.E. refers to Standard Error. Survey weights (w4w1stup1) are applied to account for students and parents’ nonresponses to enhance the external validity. This table is using percentage (0–100 scale) to describe the dichotomous variables defined in Table 7 in Appendix to ease interpretation. In the analysis, this study still uses the actual scale (0–1). In addition, the means were rounded to the nearest tenth decimal to comply with NCES restricted-data regulations. Missing data figures reported were generated from Stata 16’s “misstable” command prior to imputation
  2. Source: U.S. Department of Education, National Center for Education Statistics, High School Longitudinal Study of 2009 (HSLS: 09), Base Year, Student Survey, 2009; Parent Survey, 2009; Second Follow-up, Student Survey, 2016; Common Core of Data 2005–06, Private School Survey, 2005