Skip to main content
. 2022 Mar 19;58(1):17–23. doi: 10.1007/s00127-022-02272-4

Table 2.

Changes in predicted probabilities for endorsing continuum beliefs, unfamiliarity, and incomprehensibility for depression or schizophrenia

Statement (N) Response category Depression Schizophrenia
2011 2020 Change 95% CI 2011 2020 Change 95% CI
Continuum belief (n = 5416) Agree 42.5 46.3 3.9* 0.1, 7.7 26.1 19.8 6.2*** 9.5, 3.0
Undecided 32.7 28.8 4.0* 7.5, 0.4 24.6 25.3 0.7 − 2.6, 4.0
Disagree 24.8 24.9 0.1 − 3.2, 3.4 49.4 54.9 5.5*** 1.7, 9.3
Unfamiliar (n = 5405) Agree 26.7 23.2 3.5* 6.8, 0.2 51.9 48.4 − 3.6 − 7.4, 0.3
Undecided 29.2 23.1 6.1*** 9.4, 2.7 24.8 27.8 2.9 − 0.4, 6.3
Disagree 44.1 53.8 9.6*** 5.8, 13.4 23.2 23.9 0.6 − 2.6, 3.9
Incomprehensible (n = 5399) Agree 27.0 20.3 6.8*** 10.0, 3.5 38.3 38.1 − 0.2 − 3.9, 3.5
Undecided 25.7 23.1 − 2.6 − 5.9, 0.7 29.1 25.0 4.1* 7.5, 0.7
Disagree 47.3 56.7 9.4*** 5.5, 13.2 32.6 36.9 4.3* 0.6, 7.9

Percentages of response categories and their discrete change from 2011 to 2020 were predicted using a multinomial logistic regression with a period, vignette, interaction period*vignette, and gender of the vignette, as well as age, gender, and educational attainment of the respondent as predictors. All covariates were evaluated at their means of the combined sample. Statistically significant differences are highlighted in bold font. N sample size

*p < 0.05, ***p < 0.001