Skip to main content

Table 7 Mixed-effects logistic regression model used to predict use of lateral reading and correct trustworthiness conclusion on each problem (N = 230)

From: Improving college students’ fact-checking strategies through lateral reading instruction in a general education civics course

Predictor variables B (SE) Odds ratio (95% CI) z X2
Intercept − 8.62 (1.19) 0.00 (0.00, 0.00) − 7.24***
Media literacy accuracy at pretest 1.07 (0.80) 2.92 (0.61, 14.09) 1.33 1.83
Lateral reading at pretest (No = 0) 1.22 (0.61) 3.39 (1.02, 11.24) 1.99* 4.07*
Instructor (Instructor 1 = 0)a 10.50*
 Instructor 2 0.68 (0.51) 1.98 (0.73, 5.37) 1.33
 Instructor 3 0.22 (0.48) 1.25 (0.49, 3.18) 0.47
 Instructor 4 1.41 (0.51) 4.10 (1.52, 11.08) 2.78**
Problem type (sourcing evidence = 0)a 35.60***
 Clickbait science and medical disinformation 0.79 (0.38) 2.19 (1.03, 4.65) 2.05*
 Fake news 2.00 (0.39) 7.40 (3.47, 15.77) 5.19***
 Photographic evidence 1.13 (0.38) 3.09 (1.47, 6.50) 2.97**
Condition (Control = 0) 3.59 (0.84) 36.08 (7.02, 185.48) 4.29*** 25.10***
Number of assignments attempted 0.59 (0.21) 1.81 (1.20, 2.72) 2.85** 8.54**
  1. For instructor, post hoc comparisons with Tukey adjustment for multiple comparisons indicated that instructor 4’s students were more likely to read laterally and make a correct conclusion than instructor 1’s students (p = .028) and instructor 3’s students (p = .033). For problem type, post hoc comparisons with Tukey adjustment for multiple comparisons indicated that students were more likely to read laterally and make a correct conclusion on Fake News than Sourcing Evidence (p < .001), Clickbait Science and Medical Disinformation (p < .001), and Photo Evidence (p = .018). Students were also more likely to read laterally and correctly assess Photo Evidence than Sourcing Evidence (p = .016)
  2. p < .06, *p < .05, **p < .01, ***p < .001
  3. aBaselines set based on the lowest number of problems read laterally and correctly assessed at posttest