Skip to main content

Table 2 Results of logistic regression models predicting female students’ intentions to major in STEM Fields

From: Who is a scientist? The relationship between counter-stereotypical beliefs about scientists and the STEM major intentions of Black and Latinx male and female students

Variables

Biological sciences

Physical sciences

Math

Computer Science

Engineering

Model 1

Model 2

Model 3

Model 4

Model 5

Model 6

Model 7

Model 8

Model 9

Model 10

Model 11

Model 12

Model 13

Model 14

Model 15

Baseline

Full

Full + interaction

Baseline

Full

Full + interaction

Baseline

Full

Full + interaction

Baseline

Full

Full + interaction

Baseline

Full

Full + interaction

Counter-Stereotypical Beliefs about Scientists

0.869*

0.584

− 0.493

0.498

0.206

0.389

0.516

0.419

− 0.303

0.871*

0.953*

0.659

1.318***

1.249**

1.610

(0.409)

(0.424)

(0.911)

(0.408)

(0.428)

(1.042)

(0.330)

(0.343)

(0.865)

(0.373)

(0.385)

(0.950)

(0.392)

(0.404)

(1.084)

Interaction

 Beliefs × Latinx (ref = beliefs × Black)

  

1.346

  

− 0.218

  

0.844

  

0.347

  

− 0.418

  

(1.018)

  

(1.130)

  

(0.933)

  

(1.028)

  

(1.159)

Controls

 Latinx (ref = Black)

 

− 0.134

− 0.986

 

0.028

0.168

 

0.159

− 0.367

 

0.045

− 0.177

 

0.276

0.560

 

(0.290)

(0.692)

 

(0.306)

(0.792)

 

(0.256)

(0.627)

 

(0.278)

(0.712)

 

(0.295)

(0.848)

 Mother’s highest level of education

 

0.285

0.301

 

− 0.141

− 0.144

 

− 0.456

− 0.444

 

0.071

0.079

 

0.023

0.013

 

(0.314)

(0.313)

 

(0.346)

(0.347)

 

(0.299)

(0.299)

 

(0.305)

(0.305)

 

(0.313)

(0.315)

 Standardized 8th grade math test score

 

0.101

0.102

 

0.011

0.010

 

0.044

0.044

 

− 0.070

− 0.071

 

− 0.057

− 0.057

 

(0.113)

(0.113)

 

(0.113)

(0.113)

 

(0.093)

(0.094)

 

(0.103)

(0.103)

 

(0.104)

(0.104)

 8th grade high self-perception of science performance

 

0.654**

0.668**

 

0.883***

0.881***

 

0.587**

0.597**

 

− 0.285

− 0.281

 

0.326

0.323

 

(0.245)

(0.245)

 

(0.247)

(0.247)

 

(0.210)

(0.210)

 

(0.247)

(0.247)

 

(0.236)

(0.237)

 Constant

− 2.150***

− 2.134***

− 1.447*

− 1.970***

− 2.098***

− 2.217**

− 1.238***

− 1.424***

− 0.970

− 1.858***

− 1.882***

− 1.691*

− 2.218***

− 2.507***

− 2.755***

(0.271)

(0.387)

(0.623)

(0.264)

(0.396)

(0.741)

(0.211)

(0.323)

(0.585)

(0.245)

(0.358)

(0.662)

(0.266)

(0.389)

(0.800)

Model fit statistics

 Hosmer-Lemeshow goodness-of-fit test

  Chi-squared statistic

4.87

1.87

4.93

11.38

12.89

15.46

0.67

8.25

9.14

0.78

5.49

3.52

4.70

7.41

3.88

  df

3

8

8

3

8

8

3

8

8

3

8

8

3

8

8

  p-value

0.1819

0.9848

0.7645

0.0098**

0.1157

0.0509

0.8796

0.4097

0.3305

0.8538

0.7039

0.8975

0.1950

0.4931

0.8680

 Fit measures to compare models

  AIC

508.66

506.71

506.99

501.77

497.32

499.29

684.08

681.89

683.08

582.54

588.61

590.50

553.93

558.98

560.85

  BIC

517.37

532.84

537.48

510.49

523.46

529.78

692.79

708.03

713.57

591.26

614.75

620.99

562.64

585.12

591.34

 Likelihood ratio test (Ref: full model)

  Chi-squared statistic

9.95

 

1.72

12.45

 

0.04

10.19

 

0.81

1.93

 

0.11

2.95

 

0.13

  df

4

 

1

4

 

1

4

 

1

4

 

1

4

 

1

  p-value

0.0412*

 

0.1897

0.0143*

 

0.8466

0.0374*

 

0.3683

0.7485

 

0.7372

0.5657

 

0.7163

  1. Coefficients are from single-level logistic regression models, N = 576 female students; robust standard errors are in parentheses
  2. *** p < 0.001, ** p <0.01, * p <0.05