Table 1.
Linear regression analysis for predictors of stress in college students during the COVID-19 pandemic according to cortisol levels.
Parameter | Beta | Std. Error | 95% Wald Confidence Interval | Hypothesis Test | |||
---|---|---|---|---|---|---|---|
Lower | Upper | Wald Chi-Square | Sig. | ||||
(Intercept) | .631 | .2724 | .097 | 1.165 | 5.361 | .021 | |
Age | .006 | .0106 | −0.015 | .026 | .295 | .587 | |
Gender | |||||||
Male# | −0.144 | .0584 | −0.259 | −0.030 | 6.090 | .014* | |
Female§ | |||||||
Income | |||||||
< 2090.00 Brazilian reais/month# |
−0.007 | .0494 | −0.103 | .090 | .018 | .893 | |
> 2090.00 Brazilian reais/month§ | |||||||
Number of people they live with | .010 | .0180 | −0.025 | .046 | .333 | .564 | |
Academic semester | −0.030 | .0152 | −0.060 | −0.001 | 4.014 | .045 | |
Social distancing | |||||||
No# | −0.060 | .0871 | −0.230 | .111 | .469 | .494 | |
Yes§ | |||||||
Physical exercise | |||||||
No# | −0.027 | .0479 | −0.121 | .067 | .310 | .578 | |
Yes§ | |||||||
Chronotype (HO-MEQ) | .005 | .0023 | .001 | .010 | 5.944 | .015 | |
Sleep quality (PSQI) | −0.018 | .0080 | −0.034 | −0.002 | 4.990 | .025 | |
Anxious state due to pandemic | |||||||
Never to sometimes# | −0.012 | .0509 | −0.112 | .088 | .054 | .816 | |
Often to very often§ | |||||||
Sadness state due to pandemic | |||||||
Never to sometimes# | −0.200 | .0879 | −0.372 | −0.028 | 5.170 | .023* | |
Often to very often§ | |||||||
Study moment | |||||||
Before remote classes# | .010 | .0477 | −0.084 | .103 | .043 | .835 | |
During remote classes§ | |||||||
(Scale) | .091 | .0096 | .074 | .112 |
N = 177. Std.= Standard; df= degree of freedom; Sig.= significance; #= dichotomous covariate analyzed as 0; §= dichotomous covariate analyzed as 1; HO-MEQ= Horne & Ostberg Morningness-Eveningness Questionnaire (higher scores of this instrument indicate more morning chronotype, while lower scores indicate more evening chronotype); PSQI= Pittsburgh Sleep Quality Index (higher scores of this instrument indicate lower sleep quality). The analysis was performed using generalized linear models.
p<0.05.