Page 3 of 12
Lietal. BMC Psychology (2022) 10:125
something that they usually encounter during their edu-
cational career. Learning burnout is mainly the internal
threat giving rise to college students’ mental health prob
-
lems and unhealthy behavioral habits. Not only is learn-
ing burnout defined as a negative emotional state but
also does it relate to psychological distress [19, 20]. e
Chinese students who were haunted in the learning burn
-
out usually failed the academic performance as well [21].
As a result, they would have to bear the enormous pres
-
sure from outside and inside, followed by the high levels
of psychological distress and the increased inclinations of
smoking.
However, due to a variety of factors including learning
burnout and mental health problems might contribute to
the inclination of cigarette smoking in theory, the influ
-
ence of more possible variables would be explored in
this article. is article aimed at building a multivariate
logistic model to predict the smoking behaviors of college
students based on a wide range of independent variables
such as age, gender, majors, socioeconomic status (SES),
family environment (FE), peer influence, learning burn
-
out and psychological distress so that college students’
smoking behaviors could be accurately predicted and
intervened. Among all these variables, learning burnout
and psychological distress were expected to be strongly
associated with smoking behaviors. erefore, we will
explore the relationships among learning burnout, psy
-
chological distress and smoking behavior through empir-
ical research, discover the possible mediation mechanism
behind these variables and figure out the predictive
power of different variables on smoking behavior. For
these purposes, we further made three assumptions as
follows.
Hypothesis 1: Various demographic variables would predict
smoking behavior
Previous literature had confirmed that smoking behav-
ior was significantly associated with age, gender, majors,
SES, FE, peer influence [22–24]. ose people featured
with high age (versus low age), male (versus female), art
majors (versus science and engineering majors), high SES
(versus low SES), disharmonious and broken FE (versus
harmonious and complete FE) were more likely to smoke.
us, age, gender, majors, SES and FE would be the
demographic variables included in this article to verify
the findings of previous studies.
In addition, Wang et al. [24] had argued that friend
-
ship was also an inescapable factor that influence peo-
ple’s smoking decisions. Based on the data they collected,
Wang etal. [24] compared people who had the increased
number of smoking friends from 1989 to 1993 with those
who had the decreased number of smoking friends dur
-
ing the same period, they found that people with the
increased number of smoking friends were more likely to
smoke. It seems not only does the quality of the friend
-
ship matter, but also what kind of friends people usually
choose to make may to some extent determine the smok
-
ing behaviors. us, for purpose of verifying the convic-
tion that the type of friends college students chose to
make would affect their smoking behaviors, the predic
-
tive power of peer influence on smoking behavior was
evaluated by setting up an item like this: do you have any
friends who usually smoke around you.
Hypothesis 2: Learning burnout andits dierent dimensions
signicantly aect smoking behavior
According to the theory of planned behavior (TPB)
that was first come up with by Ajzen in 1988, perceived
behavioral control (PBC), subjective norm (SN) and atti
-
tude, were constructed as the essential factors to predict
individual’s specific intention and behavior [25]. PBC
refers to people’s perception of the ease or difficulty of
performing the behavior of interest. It was treated as the
concept of self-efficacy which was first proposed by Ban
-
dura and his associates within a general framework [26];
Correlated with PBC in the TPB model, SN is a social fac
-
tor that refers to the perceived social pressure to perform
or not to perform the behavior, and attitude refers to the
degree to which a person has a favorable or unfavorable
appraisal of the behavior [26]. It was convinced that all of
these constructs in TPB model were available to predict
a variety of health behaviors in many specific contexts
[27]. In term of smoking behavior, TPB had already been
confirmed to be the effective model to predict smoking
behavior with all of its contents including PBC, SN and
attitude significantly related to smoking behavior [28].
Although TPB was useful to predict smoking behav
-
ior, the effect sizes between TPB constructs and smoking
behavior were relatively small [29]. As Mohiyeddini etal.
[30] have argued, while TPB commonly shows high pre
-
dictive power with respect to intention, it often falls short
in the prediction of behavior, the key role in bridging the
intention-behavior gap is emotion. In this sense, emotion
should be added as the antecedent variable of smoking
behavior in the TPB model.
Following this clue, we assumed that learning burnout
as an emotional factor might relate to smoking behav
-
ior closely. Ling etal. [31] have concluded that learning
burnout has three dimensions: emotional exhaustion,
cynicism and learning efficacy. Emotional exhaustion
refers to students’ feelings of tiredness after learning.
e individual who undergoes emotional exhaustion
will excessively consume the inner cognitive resource
and thus perform a sense of fatigue. He or she will even
-
tually exhibit no extra enthusiasm and energy on learn-
ing and possibly other behaviors such as smoking. So