Fig. 5
From: Who are those random responders on your survey? The case of the TIMSS 2015 student questionnaire
Note The black diamond represents the confidence interval around the estimated average log odds ratio across educational systems, and the whiskers extending the diamond define the corresponding prediction interval. The gray dashed vertical line is drawn at log(OR) = 0, corresponding to independence between the covariate and the random responder classification. For the estimates per system, see Appendix 1: Fig. 10 and Tables 1 and 2. Number of Books at Home is coded 0 = None or very few (0–10 books)/1 = Enough to fill one shelf (11–25 books)/2 = Enough to fill one bookcase (26–100 books)/3 = Enough to fill two bookcases (101–200 books)/4 = Enough to fill three or more bookcases (more than 200 books), hence a positive/negative log(OR) indicates that the odds of having been classified as a random responder is higher for students who reported having more/fewer books at home. Results are reported for six scales in the TIMSS 2015 student questionnaire measuring three types of students’ engagement and attitudes toward Mathematics and Science
Meta-analytic confidence and prediction intervals for the odds of having been classified as a Random Responder as a function of the student’s Number of Books at Home.