Health Promotion International, 2017, 1–8

doi: 10.1093/heapro/dax080
Article

Effects of a healthy-eater self-schema and

nutrition literacy on healthy-eating behaviors
among Taiwanese college students
Chia-Kuei Lee1, Li-Ling Liao2,*, I-Ju Lai3, and Li-Chun Chang4
1
Department of Nursing, College of Medicine, National Cheng Kung University, Tainan 70101, Taiwan,
2
Department of Health Management, I-Shou University, Kaohsiung City 82445, Taiwan, 3Department of
Nutrition, I-Shou University, Kaohsiung City 82445, Taiwan and 4Department of Nursing, Chang Gung
University of Science and Technology, Tao-Yuan 33303, Taiwan
*Corresponding author. E-mail: hililin@isu.edu.tw

Summary
Unhealthy eating behaviors contribute to obesity and chronic illness. This study examined the relative
contributions of a healthy-eater self-schema (a self-conception as a healthy eater) and nutrition liter-
acy on healthy-eating behaviors and whether nutrition literacy was a mediator among Taiwanese col-
lege students. A total of 1216 undergraduate students from six universities in Taiwan participated in
the study from April to June 2016. Healthy-eating behaviors, nutrition literacy, healthy-eater self-
schema and known determinants of eating behaviors (e.g. nutrition-related information, health status,
nutrition knowledge needs, sex, year in college and residence) were measured by a self-report ques-
tionnaire. A hierarchical multiple regression and mediation analysis were conducted with the known
determinants of eating behaviors as covariates. Results showed that a healthy-eater self-schema and
nutrition literacy explained 9% and 12% of the variance in healthy-eating behaviors, respectively, and
both had unique effects on healthy-eating behaviors. The effect of a healthy-eater self-schema on
healthy-eating behaviors was partially mediated through nutrition literacy. Findings suggest that both
a healthy-eater self-schema and nutrition literacy should be considered when promoting healthy-
eating behaviors. Additionally, nutrition literacy interventions should be tailored to the healthy-eater
self-schema status and emphasize the personal relevance of being a healthy-eater to improve the in-
tervention’s effectiveness.

Key words: identity, nutrition, literacy, college students, eating behavior

INTRODUCTION (Anding et al., 2001; Hiza and Gerrior, 2002; Racette

College is a critical period of lifestyle change and for the et al., 2005). This contributes to weight problems, poor
formation of health behaviors, including dietary pat- nutrition and negative health consequences (Hu et al.,
terns. College students often display unhealthy-eating 2000; Van Duyn and Pivonka, 2000; Bazzano et al.,
behaviors and consume too much fat, sodium and calo- 2002; Racette et al., 2005). The American
ries, while consuming too few fruits and vegetables College Health Association–National College Health

C The Author 2017. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com
V

Downloaded from https://academic.oup.com/heapro/advance-article-abstract/doi/10.1093/heapro/dax080/4628149

by Mount Royal University user
on 24 January 2018
 2 C. K. Lee et al.

Assessment, 2015 report revealed that approximately an extension of health literacy (IOM, 2004) that con-
97% of college students did not consume five or more cerns dietary performance, which reflects the compe-
servings of fruits and vegetables (the daily recommended tence of healthy-eating and has been shown to influence
amount) and 39% were overweight/obese (American healthy-eating behaviors. Cha and colleagues (2014)
College Health Association, 2016). In Taiwan, accord- found that young adults with low health literacy were
ing to a 2011 national survey, 17% of college students less likely to use food labels and had lower dietary qual-
were overweight/obese and less than 10% of college stu- ity than those with high health literacy. Another study
dents met the daily dietary recommendations (Lin et al., showed that higher health literacy was associated with
2011). Moreover, college students’ eating behaviors higher healthy-eating index scores and less consumption
commonly varied by sex, residence, year in college and of sugar-sweetened beverages (Zoellner et al., 2011).
perceived health status (Brevard and Ricketts, 1996; Although a growing body of research has focused on the
Wardle et al., 2004; Driskell et al., 2005; Racette et al., development and implementation of health literacy in-
2005; Davy et al., 2006; Kim et al., 2008; Papier et al., terventions in promoting health behaviors (Taggart
2015; Yahia et al., 2016). et al., 2012; Wei et al., 2013), these interventions have
One’s self-conception (referred to as self-schema) has focused on enhancing people’s abilities—emphasizing
become increasingly recognized for its importance in the improvement of relevant knowledge, attitude and
regulating health behaviors, including eating behaviors skills—but did not take into account the individual dif-
(Stein and Markus, 1996; Kendzierski, 2007; Wheeler ferences in people’s cognitive structures about the self.
et al., 2007). A self-schema is an elaborated knowledge Further exploration of how nutrition literacy functions
structure about the self that contains the relevant seman- with a healthy-eater self-schema would contribute to the
tic (e.g. abstraction or generalization memory), episodic development of effective intervention strategies.
(e.g. memory of specific episodes) and procedural Considering competence (i.e. literacy) in a domain
knowledge (e.g. action-based memory including skills may require a domain-specific self-schema to facilitate
and strategies) (Kihlstrom et al., 2005; Ghosh and the development and use of relevant abilities in a given
Gilboa, 2014). A self-schema in a specific content do- domain (Cross and Markus, 1994), a healthy-eater self-
main reflects not only self-descriptiveness, but also an schema may be contingent on strengthening nutrition lit-
important aspect of the self, which enhances the process- eracy by enhancing learning and encoding nutrition
ing of schema-congruent information and motivates knowledge and skills, which promotes adherence to
schema-congruent behaviors (Kendzierski, 1990; healthy-eating behaviors. Accordingly, this study ex-
Markus, 1977; Markus et al., 1987; Stein and Corte, tended previous work about the role of the self in
2007). Each individual has a unique collection of healthy-eating behaviors by incorporating nutrition lit-
self-schemas in different content domains, which are de- eracy. We aimed to determine (1) the relative contribu-
rived from personal experiences, values and social con- tions of a healthy-eater self-schema and nutrition
texts (Markus, 1977; Markus and Wurf, 1987). literacy on healthy-eating behaviors and (2) whether nu-
Researchers have demonstrated the link between a trition literacy mediates the effects of a healthy-eater
healthy-eater self-schema and healthier dietary intake self-schema on healthy-eating behaviors among under-
among female college students and middle-aged working graduate students.
adults: those with a healthy-eater self-schema consumed
more carbohydrates (e.g. fiber, fruits and vegetables),
less total fat (or cholesterol) and less junk food than
METHODS
those without a healthy-eater self-schema (Kendzierski
and Costello, 2004; Noureddine and Stein, 2009; Allom Participants and procedure
and Mullan, 2012; Holub et al., 2012). Though a A convenience sample of participants was drawn from
healthy-eater self-schema has been identified as playing three national and three private 4-year comprehensive
an important role in dietary intake/behaviors, self- universities in northern, central and southern Taiwan
schema approaches have not commonly been applied in (one national and one private university in each region).
healthy-eating intervention studies. To obtain a diverse group of participants and avoid the
In addition to self-schemas, researchers have sug- influence of students’ majors on nutrition knowledge
gested that health literacy is an important factor in and eating behaviors (Hong et al., 2016), 220 under-
the context of nutrition behaviors or dietary intake graduate students were recruited from each school with
(Silk et al., 2008; Cha et al., 2014). Nutrition literacy is a maximum of 45 students from each academic

Downloaded from https://academic.oup.com/heapro/advance-article-abstract/doi/10.1093/heapro/dax080/4628149

by Mount Royal University user
on 24 January 2018
 Effects of a healthy-eater self-schema and nutrition literacy on healthy-eating behaviors 3

discipline. In addition, in each school, the total number situation; 1 item), ‘appraise’ (i.e. having the capacity to
of students with medical, public health and nutrition re- judge and assess nutrition information in terms of per-
lated majors were limited to 45 students to avoid sample sonal needs; 2 items) and ‘apply’ (i.e. having the capacity
bias, because students from those majors may have to apply nutrition information to daily life in order to
higher awareness of healthy eating. achieve a healthy diet; 1 item), based on the definition of
The trained research assistants approached students nutrition literacy (Nielsen-Bohlman et al., 2004; Silk
on campus wherever they could be found (e.g. class- et al., 2008; Aihara and Minai, 2011). Response options
rooms, school hallways, or college dorms), provided in- ranged from 1 (very difficult) to 4 (very easy). A sum
formation about the study, and invited them to score was computed, with a higher score reflecting
participate (April–June, 2016). Upon obtaining written higher nutrition literacy. In the pilot testing with 492
informed consent, students were asked to complete a college students, the Cronbach’s alpha was .87 and aver-
self-administered questionnaire individually, which con- age CVI was 1. Additionally, in a subsample (N ¼ 119)
tained measures of nutrition literacy, healthy-eating be- of these college students, the ICC of the 2-week test-
haviors, self-schemas, known determinants of retest reliability was 0.53 (p < 0.001). The Cronbach’s
eating behaviors and demographic information. The alpha was 0.85 in the current study.
questionnaire took approximately 30 min to complete. The healthy-eater self-schema was rated on an
Participants were also informed that they could with- 11-point scale addressing three key phrases: healthy
draw at any stage without adverse personal conse- eater, someone who eats in a nutritious manner and
quences. A small gift was provided in compensation someone who is careful about what they eat, according
after the successful completion of the questionnaire. to self-descriptiveness and importance separately
All study procedures were approved by the university’s (Kendzierski and Costello, 2004). A rating of at least
Human Research Ethics Committee. Of the question- two of the three descriptors as both very descriptive
naires distributed (N ¼ 1320), 1216 completed (8  11 points) and very important (8  11 points) was
questionnaires that contained sufficient information considered to indicate the presence of a healthy-eater
for primary measures were returned (92.1% valid self-schema. The appropriate psychometric properties
response rate). for this measure have been demonstrated in college stu-
dents and middle-aged working adults (Kendzierski and
Measures Costello, 2004; Noureddine and Stein, 2009).
Healthy-eating behaviors were measured by a 13-item Known determinants of eating behaviors were mea-
frequency rating scale on healthy eating practices, sured to control for alternative explanations of the
including a balanced diet (8 items), processed food relationship of nutrition literacy and a healthy-eater self-
consumption (1 item), nutrition label usage (3 items) schema with healthy-eating behaviors. Nutrition-related
and healthy-eating choices (1 item). Response options information/education was measured by a question,
ranged from 1 (never) to 5 (always). A sum score was ‘How frequently do you receive nutrition-related infor-
computed, with a higher score indicating better mation from course work, outreach activities, or mass
healthy-eating behaviors. In the pilot testing with 492 media during your college years?’ Responses ranged
college students, the Cronbach’s alpha was 0.87 and from 1 (never) to 5 (always). Health status was mea-
the average content validity index (CVI) was 0.99. sured by an item on perceived current health status,
Additionally, in a random subsample (N ¼ 119) of ranging from 1 (very good) to 5 (very bad). Nutrition
these college students, the intra-class correlation coeffi- knowledge needs were measured by an item rating from
cient (ICC) of the 2-week test-retest reliability was 1 (no need) to 4 (definitely need), based on the demands
0.49 (p < 0.001). Cronbach’s alpha was 0.86 in the of nutrition-related knowledge. Sex, year in college and
current study. residence (living on or off campus during the semester)
Nutrition literacy was measured by an 8-item scale were also measured.
on capacities regarding nutritional information in five
dimensions: ‘obtain’ (i.e. having the capacity to search Statistical analysis
for, find, and acquire nutrition information; 2 items), Descriptive statistics were conducted to provide infor-
‘understand’ (i.e. having basic nutrition knowledge and mation about participants’ characteristics. T-tests and
the capacity to understand general nutrition informa- chi-square tests were used for comparisons between un-
tion; 2 items), ‘analyse’ (i.e. having the capacity to dis- dergraduate students with and without a healthy-eater
criminate and analyse nutrition information in a given self-schema. A hierarchical multiple regression with

Downloaded from https://academic.oup.com/heapro/advance-article-abstract/doi/10.1093/heapro/dax080/4628149

by Mount Royal University user
on 24 January 2018
 4 C. K. Lee et al.

Table 1: Descriptive statistics by healthy-eater self-schema groups

All With a healthy-eater Without a healthy-eater t p

self-schema self-schema

(n 5 268, 22.04%) (n 5 948, 77.96%)

Variable Mean (SD) Ranges Mean (SD) Ranges Mean (SD) Ranges

Nutrition literacy 22.41 (3.66) 8–32 23.98 (3.66) 8–32 21.97 (3.54) 11–32 –8.17 <0.001
Healthy-eating behaviors 39.97 (7.73) 17–65 45.52 (6.71) 17–65 38.36 (7.18) 17–58 –4.61 <0.001

Freq. Percent Freq. Percent Freq. Percent v2 p

Sex 13.84 <0.001

Male 559 45.97 150 26.83 409 73.17
Female 657 54.03 118 17.96 539 82.04
Residence 0.00 0.945
On campus 642 52.80 141 21.96 501 78.04
Off campus 574 47.20 127 22.13 447 77.87
Year in college 2.23 0.526
Freshman 225 18.50 43 19.11 182 80.89
Sophomore 466 38.32 107 22.96 359 77.04
Junior 324 26.64 77 23.77 247 76.23
Senior 201 16.53 41 20.40 160 79.60
Nutrition information 25.64 <0.001
Never 76 6.25 15 19.74 61 80.26
Rarely 410 33.72 69 16.83 341 83.17
Sometimes 496 40.79 105 21.17 391 78.83
Often 169 13.90 57 33.73 112 66.27
Always 65 5.35 22 33.85 43 66.15
Health status 168.37 <0.001
Very good 550 45.23 213 38.73 337 61.27
Good 220 18.09 30 13.64 190 86.36
Fair 371 30.51 21 5.66 350 94.34
Poor 71 5.84 4 5.63 67 94.37
Very bad 4 0.33 0 0.00 4 100.00
Nutrition knowledge needs 10.13 0.017
No need 42 3.45 7 16.67 35 83.33
Not really need 292 24.01 56 19.18 236 80.82
Need 722 59.38 155 21.47 567 78.53
Definitely need 160 13.16 50 31.25 110 68.75

three steps (step 1: covariates of the known determinants eating behaviors (i.e. nutrition-related information,
of healthy-eating behaviors; step 2: healthy-eater self- health status, nutrition knowledge needs, sex, year in
schema and step 3: nutrition literacy) was used to college and residence) as covariates.
estimate the unique predictions of a healthy-eater self-
schema and nutrition literacy. Then, mediation analysis
with a bootstrapping test (Preacher and Hayes, 2008)
RESULTS
was conducted to determine whether nutrition literacy
mediated the effects of a healthy-eater self-schema on Participants’ characteristics
healthy-eating behaviors. The significant indirect effect Among the 1216 undergraduate students, 46.0% were
was identified based on a 1000 times resampling with a male. Table 1 shows the descriptive statistics for study
95% bias-corrected confidence interval (CI). The pro- variables by healthy-eater self-schema groups. Over one-
portion of the total effect that is mediated was also com- third of the students (38.3%) were sophomores, fol-
puted. All analyses included the known determinants of lowed by 26.6% who were juniors. More than half of

Downloaded from https://academic.oup.com/heapro/advance-article-abstract/doi/10.1093/heapro/dax080/4628149

by Mount Royal University user
on 24 January 2018
 Effects of a healthy-eater self-schema and nutrition literacy on healthy-eating behaviors 5

the students (52.8%) lived on campus. Twenty-two per- the significant direct effect of a healthy-eater self-schema
cent of students were classified as having a healthy-eater on nutrition literacy (B ¼ 1.25, standard error
self-schema, with a significant sex difference (26.8% (SE) ¼ 0.25, 95% CI ¼ 0.75  1.74) and healthy-eating
male vs. 18.0% female). The average sum score of nutri- behaviors (B ¼ 4.89, SE ¼ 0.48, 95% CI ¼ 3.95  5.82).
tion literacy was 22.4 (SD ¼ 3.7), and the average sum Moreover, the indirect effect of a healthy-eater self-
score of healthy-eating behaviors was 40.0 (SD ¼ 7.7). schema on healthy-eating behaviors through nutrition
Undergraduate students who had a healthy-eater self- literacy was significant (B ¼ 0.97, SE¼ 0.23, 95%
schema reported significantly higher scores of nutrition CI ¼ 0.56  1.48), with 16.5% of the total effect that is
literacy and healthy-eating behaviors compared to those mediated.
without a healthy-eater self-schema.
In terms of the known determinants, most students
reported that they had received nutrition-related infor- DISCUSSION
mation from course work, outreach activities, or mass This study examined the relative contributions of a
media either rarely (33.7%) or sometimes (40.8%) dur- healthy-eater self-schema and nutrition literacy on
ing their college years. Approximately two-thirds of the healthy-eating behaviors among undergraduate students
students reported a good health status currently (45.2% and sought to determine whether nutrition literacy was
very good and 18.0% good), and 30.5% reported a fair a mediating variable. Consistent with existing literature,
health status. With regard to the demands for nutrition- our findings showed that a healthy-eater self-schema
related knowledge, more than 70% of the students re- and nutrition literacy had positive and unique effects on
ported that they had needs for nutrition-related healthy-eating behaviors. Since most previous studies
knowledge. Overall, undergraduate students with a were conducted in Western countries, this finding
healthy-eater self-schema were more likely to report among a Taiwanese population provides cross-cultural
more frequent reception of nutrition-related informa- evidence that a healthy-eater self-schema and nutrition
tion, a better health status, and more needs for literacy are stable predictors of healthy-eating behaviors
nutrition-related knowledge. regardless of being from a Western or an Asian culture.
Moreover, we also found that the effect of nutrition lit-
eracy on healthy-eating behaviors was stronger for un-
Healthy-eater self-schema and nutrition literacy dergraduate students with a healthy-eater self-schema
on healthy-eating behaviors compared to those without. This suggests that while
The final step of the hierarchical regression both a healthy-eater self-schema and nutrition literacy
model for healthy-eating behaviors was significant; contain relevant knowledge, skills and cognitive func-
F(8, 1207) ¼ 70.06, p < 0.001, R2 ¼ 0.32. Results indi- tioning, they may tap into different cognitive levels in re-
cated that a healthy-eater self-schema explained 9% lation to these knowledge and skills. Therefore, having
of the variance in healthy-eating behaviors; both a healthy-eater self-schema and nutrition literacy
F(1, 1208) ¼ 129.58, p < 0.001, and nutrition literacy would be more suitable to regulate dietary-related be-
explained an additional 12% of the variance in healthy- haviors than having either one alone. Accordingly, inter-
eating behaviors; F(1, 1207) ¼ 207.16, p < 0.001. ventions to improve healthy-eating behaviors may be
Partial regression coefficients are reported in Table 2. more effective if they incorporate the enhancement of
Results showed that undergraduate students who had a both the personal value of healthy eating and nutrition
healthy-eater self-schema and a higher level of nutri- literacy at the same time.
tion literacy were more likely to engage in healthy- In addition to the direct effects on healthy-eating be-
eating behaviors than others were. Figure 1 also haviors, findings also showed that the effect of a
displays the relationships between nutrition literacy healthy-eater self-schema on healthy-eating behaviors
and healthy-eating behaviors by healthy-eater self- was partially mediated through nutrition literacy,
schema groups. Healthy-eating behaviors increased though for only a small proportion of the total effect
with the improvement of nutrition literacy in both that is mediated. Moreover, a healthy-eater self-schema
groups (with and without a healthy-eater self-schema). had a direct effect on nutrition literacy. Therefore, a
The healthy-eater self-schema group had more healthy- healthy-eater self-schema enhances the competence of
eating behaviors compared to the group without a healthy-eating, nutrition literacy and in turn affects
healthy-eater self-schema. healthy-eating behaviors. Indeed, in our sample, stu-
In addition, after controlling for the known determi- dents who had a healthy-eater self-schema had higher
nants of eating behaviors, a mediation analysis showed nutrition literacy and more healthy-eating behaviors

Downloaded from https://academic.oup.com/heapro/advance-article-abstract/doi/10.1093/heapro/dax080/4628149

by Mount Royal University user
on 24 January 2018
 6 C. K. Lee et al.

Table 2: Healthy-eater self-schema and nutrition literacy on healthy-eating behaviors

Predictors B SE(B) b t p B SE(B) b t p B SE(B) b t p

Step 1
Nutrition information 1.61 0.23 0.20 6.94 <0.001 1.40 0.22 0.17 6.31 <0.001 0.72 0.21 0.09 3.42 0.001
Health status –1.82 0.21 –0.24 –8.71 <0.001 –1.02 0.21 –0.13 –4.82 <0.001 –0.58 0.20 –0.08 –2.95 0.003
Nutrition knowledge needs 0.52 0.31 0.05 1.69 0.092 0.27 0.30 0.20 0.90 0.367 0.13 0.27 0.01 0.46 0.645
Sex –0.37 0.42 –0.02 –0.89 0.374 –0.01 0.40 –0.00 –0.02 0.981 0.14 0.37 0.01 0.38 0.704
Year in college 0.03 0.22 0.00 0.15 0.884 –0.02 0.21 –0.00 –0.09 0.930 0.26 0.20 0.03 1.31 0.190
Residence 0.02 0.44 0.00 0.04 0.970 –0.12 0.42 –0.01 –0.29 0.773 –0.27 0.39 –0.02 –0.70 0.481
Step 2
Healthy-eater self-schema 5.85 0.51 0.32 11.38 <.001 4.91 0.48 0.27 10.23 <0.001
Step 3
Nutrition literacy 0.77 0.05 0.37 14.33 <0.001
R2 0.11 0.20 0.32
DR2 0.09 0.12
DF(df), p value F (1, 1208) ¼ 129.58, p < 0.001 F (1, 1207) ¼ 207.16, p < 0.001

Unlike our findings, Holub and colleagues (2012) found

that a higher proportion of female college students (25%)
were classified as having a healthy-eater self-schema than
their male counterparts (9%); however, they had a rela-
tively small number of men in their sample (n ¼ 23, 18%).
Other studies also showed that females were more likely to
report more perceived importance for being a healthy eater
or held greater beliefs in the importance of healthy diets
than males did. The belief was that this was due to not
only stereotypes that healthy foods are more feminine and
unhealthy foods are more masculine, but also that there are
different stages of change in translating these perceptions
Fig. 1: Relationships of nutrition literacy with healthy-eating be-
to actions (Wardle et al., 2004; Vartanian et al., 2007).
haviors by healthy-eater self-schema groups.
However, they did not measure a healthy-eater self-
schema. Since this is the first study to explore the healthy-
than their non-schema counterparts. This suggests that a eater self-schema in Asian countries, it is unclear whether
healthy-eater self-schema may facilitate the processing cultural context or sex stereotypes influence the develop-
of information related to healthy-eating, which enhances ment of a healthy-eater self-schema in males and females in
the learning of nutrition literacy and promotes the readi- Asia. Future research is needed to verify these findings.
ness of engaging in healthy-eating behaviors. As such, to Furthermore, this sample showed a relatively low
increase their effectiveness, nutrition literacy interven- prevalence of a healthy-eater self-schema compared to
tions should be tailored to the healthy-eater self-schema other studies (Noureddine and Stein, 2009; Allom and
status, since perceived healthy eating is important to Mullan, 2012). This may reflect the fact that the role of
oneself and may promote the encoding of nutrition liter- the self in relation to healthy eating is commonly over-
acy thus motivating healthy eating. Compared to those looked in this population as well as in the relevant inter-
with a healthy-eater self-schema, students without it ventions. Researchers have shown that college students
may need to have additional sessions or more frequent with a healthy-eater self-schema had shorter lapses in
reviews and practice of nutrition-related knowledge and healthy eating and more personal control for the causes
skills to implement healthy dietary habits (Kendzierski of lapses than those without a healthy-eater self-schema
and Costello, 2004; Kendzierski, 2007). (Holub et al., 2012). This suggests that individuals who
Moreover, we found sex differences in the prevalence internalize healthy eating as being important to the self
of the healthy-eater self-schema, as males had a signifi- can strengthen and sustain their behaviors. Therefore,
cantly higher proportion (27%) than females did (18%). individuals with a healthy-eater self-schema may be

Downloaded from https://academic.oup.com/heapro/advance-article-abstract/doi/10.1093/heapro/dax080/4628149

by Mount Royal University user
on 24 January 2018
 Effects of a healthy-eater self-schema and nutrition literacy on healthy-eating behaviors 7

more likely to persist or stabilize their healthy dietary in- should consider both a healthy-eater self-schema and
take regardless of the external circumstances. As such, nutrition literacy to enhance their effectiveness.
nutrition literacy interventions that incorporate the de-
velopment of a healthy-eater schema and enhance the
personal value of healthy eating could enhance the con- FUNDING
gruence between the capacity of literacy and adherence The authors declared receipt of the following financial
to healthy-eating behaviors. The possible strategies to support for the research, authorship and/or publication
develop a healthy-eater schema may include working of this article: This work was supported by the Taiwan
with individuals to stimulate interest in healthy eating, Ministry of Science and Technology in 2015 [grant num-
providing opportunities to think about the possibilities ber MOST104-2511-S-214-001].
of and engaging in healthy-eating practices, directing the
reflection to the risks and benefits of their own eating
practices, setting up the short-term and long-term goals REFERENCES
of healthy eating, developing a plan of action that in- Aihara, Y. and Minai, J. (2011) Barriers and catalysts of nutri-
cludes specific tasks and a timeline to achieve healthy- tion literacy among elderly Japanese people. Health
eating goals. In addition, future research is needed to Promotion International, 26, 421–431.
examine whether the healthy-eater self-schema status Allom, V. and Mullan, B. (2012) Self-regulation versus habit:
changes throughout nutrition literacy interventions and the influence of self-schema on fruit and vegetable consump-
tion. Psychology and Health, 27, 7–24.
whether an existing healthy-eater self-schema enhances
American College Health Association. (2016) American College
the effectiveness of nutrition literacy interventions.
Health Association–National College Health Assessment II:
Reference Group Executive Summary Fall 2015.
Undergraduate students. Hanover, MD, American College
Limitations
Health Association.
There are a few limitations of the current study that Anding, J. D., Suminski, R. R. and Boss, L. (2001) Dietary in-
should be noted. Due to the cross-sectional study design, take, body mass index, exercise, and alcohol: Are college
causality could not be established. Future longitudinal women following the dietary guidelines for Americans?
and intervention-based studies are needed to verify the Journal of American College Health, 49, 167–171.
prospective relationships among healthy-eater self- Bazzano, L. A., He, J., Ogden, L. G., Loria, C. M., Vupputuri,
schema, nutrition literacy and healthy-eating behaviors. S., Myers, L. et al. (2002) Fruit and vegetable intake and risk
Moreover, the measures of healthy-eating behavior in this of cardiovascular disease in US adults: the first National
study did not cover detailed information of daily healthy- Health and Nutrition Examination Survey Epidemiologic
Follow-up Study. The American Journal of Clinical
eating practices. Future studies that use a daily diary as-
Nutrition, 76, 93–99.
sessment or a more detailed food frequency questionnaire
Brevard, P. B. and Ricketts, C. D. (1996) Residence of college stu-
to examine eating behaviors may be beneficial. Moreover, dents affects dietary intake, physical activity, and serum lipid
prospective or real-time behavioral record that reduces re- levels. Journal of the American Dietetic Association, 96, 35–38.
call bias may be more accurate in estimating eating behav- Cha, E. S., Kim, K. H., Lerner, H. M., Dawkins, C. R., Bello, M.
iors. In addition, this study did not include the covariates K., Umpierrez, G. et al. (2014) Health literacy, self-efficacy,
of students’ majors and their school and life events (e.g. food label use, and diet in young adults. American Journal
midterm examinations, clinical placements, club activities, of Health Behavior, 38, 331–339.
competitions, etc.), which may influence their eating be- Cross, S. and Markus, H. (1994) Self-schemas, possible selves,
haviors. Future research needs to take into account those and competent performance. Journal of Educational
Psychology, 86, 423–438.
factors in order to clarify the findings.
Davy, S. R., Benes, B. A. and Driskell, J. A. (2006) Sex differ-
ences in dieting trends, eating habits, and nutrition beliefs of
a group of midwestern college students. Journal of the
CONCLUSIONS American Dietetic Association, 106, 1673–1677.
Driskell, J. A., Kim, Y.-N. and Goebel, K. J. (2005) Few differ-
These results provide evidence for the unique effects of a
ences found in the typical eating and physical activity habits
healthy-eater self-schema and nutrition literacy on of lower-level and upper-level university students. Journal of
healthy-eating behaviors. In addition, nutrition literacy the American Dietetic Association, 105, 798–801.
partially mediated the effects of a healthy-eater self- Ghosh, V. E. and Gilboa, A. (2014) What is a memory schema?
schema on healthy-eating behaviors. Findings indicate A historical perspective on current neuroscience literature.
that interventions to promote healthy-eating behaviors Neuropsychologia, 53, 104–114.

Downloaded from https://academic.oup.com/heapro/advance-article-abstract/doi/10.1093/heapro/dax080/4628149

by Mount Royal University user
on 24 January 2018
 8 C. K. Lee et al.

Hiza, H. and Gerrior, S. A. (2002) Using the Interactive Healthy Papier, K., Ahmed, F., Lee, P. and Wiseman, J. (2015) Stress and
Eating Index to assess the quality of college students’ diets. dietary behaviour among first-year university students in
Family Economics and Nutrition Review, 14, 3–12. Australia: Sex differences. Nutrition, 31, 324–330.
Holub, S. C., Haney, A. M. and Roelse, H. (2012) Deconstructing Preacher, K. J. and Hayes, A. F. (2008) Asymptotic and resam-
the concept of the healthy eater self-schematic: relations to die- pling strategies for assessing and comparing indirect effects
tary intake, weight and eating cognitions. Eating Behaviors, in multiple mediator models. Behavioral Research Methods,
13, 106–111. 40, 879–891.
Hong, M. Y., Shepanski, T. L. and Gaylis, J. B. (2016) Majoring Racette, S. B., Deusinger, S. S., Strube, M. J., Highstein, G. R.
in nutrition influences BMI of female college students. and Deusinger, R. H. (2005) Weight changes, exercise, and
Journal of Nutritional Science, 5, 1–7. dietary patterns during freshman and sophomore years of
Hu, F. B., Rimm, E. B., Stampfer, M. J., Ascherio, A., Spiegelman, college. Journal of American College Health, 53, 245–251.
D. and Willett, W. C. (2000) Prospective study of major die- Silk, K. J., Sherry, J., Winn, B., Keesecker, N., Horodynski, M.
tary patterns and risk of coronary heart disease in men. The A. and Sayir, A. (2008) Increasing nutrition literacy: testing
American Journal of Clinical Nutrition, 72, 912–921. the effectiveness of print, web site, and game modalities.
IOM. (2004) Health Literacy: A Prescription to End Confusion. Journal of Nutrition Education and Behavior, 40, 3–10.
National Academies Press, Washington, DC. Stein, K. F. and Corte, C. (2007) Identity impairment and the
Kendzierski, D. (1990) Exercise self-schemata: cognitive and be- eating disorders: Content and organization of the
havioral correlates. Health Psychology, 9, 69–82. self-concept in women with anorexia nervosa and bulimia
Kendzierski, D. (2007) A self-schema approach to healthy eat- nervosa. European Eating Disorders Review, 15, 58–69.
ing. Journal of the American Psychiatric Nurses Stein, K. F. and Markus, H. (1996) The role of the self in behavioral
Association, 12, 350–357. change. Journal of Psychotherapy Integration, 6, 349–384.
Kendzierski, D. and Costello, M. C. (2004) Healthy eating Taggart, J., Williams, A., Dennis, S., Newall, A., Shortus, T.,
self-schema and nutrition behavior. Journal of Applied Zwar, N. et al. (2012) A systematic review of interventions
Social Psychology, 34, 2437–2451. in primary care to improve health literacy for chronic dis-
Kihlstrom, J. F., Beer, J. S. and Klein, S. B. (2005) Self and iden- ease behavioral risk factors. BMC Family Practice, 13, 1.
tity as memory. In Leary, M. R., MacDonald, G. and Van Duyn, M. A. S. and Pivonka, E. (2000) Overview of the
Tangney, J. P. (eds), Handbook of Self and Identity. health benefits of fruit and vegetable consumption for the di-
Guilford Press, New York, pp. 68–90. etetics professional: selected literature. Journal of the
Kim, M. J., Lim, Y. R. and Kwak, H. K. (2008) Dietary behav- American Dietetic Association, 100, 1511–1521.
iors and body image recognition of college students accord- Vartanian, L. R., Herman, C. P. and Polivy, J. (2007)
ing to the self-rated health condition. Nutrition Research Consumption stereotypes and impression management:
and Practice, 2, 107–113. How you are what you eat. Appetite, 48, 265–277.
Lin, W., Hang, J. M., Wang, R. L., Lai, I. J. and Yang, H. C. Wardle, J. P., Haase, A. M. P., Steptoe, A. D., Nillapun, M. P.,
(2011) Investigation of dietary environment of universities Jonwutiwes, K. P. and Bellisie, F. P. (2004) Gender differ-
and eating patterns of college students [in Chinese] (pp. ences in food choice: the contribution of health beliefs and
1-205): Health Promotion Administration, Ministry of dieting. Annals of Behavioral Medicine, 27, 107–116.
Health and Welfare. Wei, Y., Hayden, J. A., Kutcher, S., Zygmunt, A. and McGrath,
Markus, H. (1977) Self-schemata and processing information P. (2013) The effectiveness of school mental health literacy
about the self. Journal of Personality and Social Psychology, programs to address knowledge, attitudes and help seeking
35, 63–78. among youth. Early Intervention in Psychiatry, 7, 109–121.
Markus, H., Hamill, R. and Sentis, K. P. (1987) Thinking fat: Wheeler, S. C., DeMarree, K. G. and Petty, R. E. (2007)
Self-schemas for body weight and the processing of weight Understanding the role of the self in prime-to-behavior ef-
relevant information. Journal of Applied Social Psychology, fects: The active-self account. Personality and Social
17, 50–71. Psychology Review, 11, 234–261.
Markus, H. and Wurf, E. (1987) The dynamic self-concept: a so- Yahia, N., Wang, D., Rapley, M. and Dey, R. (2016)
cial psychological perspective. Annual Review of Assessment of weight status, dietary habits and beliefs, phys-
Psychology, 38, 299–337. ical activity, and nutritional knowledge among university
Nielsen-Bohlman, L., Panzer, A. and Kindig, D. (2004). Health students. Perspectives in Public Health, 136, 231–244.
Literacy: A Prescription to End Confusion [Committee on Zoellner, J., You, W., Connell, C., Smith-Ray, R. L., Allen, K.,
Health Literacy; Board on Neuroscience and Behavioral Tucker, K. L. et al. (2011) Health literacy is associated with
Health]. National Academies Press, Washington, DC. healthy eating index scores and sugar-sweetened beverage
Noureddine, S. and Stein, K. (2009) Healthy-eater self-schema intake: Findings from the rural lower Mississippi Delta.
and dietary intake. Western Journal of Nursing Research, Journal of the American Dietetic Association, 111,
31, 201–218. 1012–1020.

Downloaded from https://academic.oup.com/heapro/advance-article-abstract/doi/10.1093/heapro/dax080/4628149

by Mount Royal University user
on 24 January 2018