Computers & Education 94 (2016) 204e214

Contents lists available at ScienceDirect

Computers & Education

journal homepage: www.elsevier.com/locate/compedu

The moderating effects of parenting styles on the relation

between the internet attitudes and internet behaviors of
high-school students in Taiwan
Hui-Lien Chou, Chien Chou, Chao-Hsiu Chen*
Institute of Education, National Chiao Tung University, 1001 Ta-Hsueh Rd., Hsinchu 30010, Taiwan, ROC

a r t i c l e i n f o a b s t r a c t

Article history: The purpose of this study is to investigate how parenting styles moderate the relation
Received 6 June 2015 between the Internet attitudes and behaviors of high-school-aged students in Taiwan. We
Received in revised form 25 November 2015 review the literature on studies of parenting styles and Internet attitudes, and their re-
Accepted 30 November 2015
lations between Internet behaviors. We employed Latent Class Analysis to cluster the
Available online 15 December 2015
child-reported parenting styles. In this regard, we found that there exist three types of
parenting styles in Taiwan. We in turn implemented the 2nd-order confirmatory factor
Keywords:
analysis to validate the 6-T internet attitude model. The so-called 6-T internet attitude
Country-specific developments
Media in education
model is to explain the Internet attitude by the multi-dimensions such as Tool, Telephone,
Secondary education Toy, Territory, Treasure and Trade. It provides the basis on which we conducted the
structural regression modeling. Three kinds of Internet behavior “online social compen-
sation seeking”, “online game addiction” and “information security behavior” are analyzed
in our research. We used the structural regression modeling to incorporate the 2nd-order
confirmatory factor analysis and path analysis so that we can tell the moderating effects of
parenting styles on the relation between Internet attitudes and Internet behavior by multi-
group analysis. The result revealed that parenting styles have moderation effect on the
relation between Internet attitude and “information security behavior”. Implications of the
findings were discussed followed by the statistical analysis.
© 2015 Elsevier Ltd. All rights reserved.

1. Introduction

Internet use permeates all age groups and all aspects of life. It is not unusual to see a baby swiping a hand-held device in
his parents' company. It is also common to see the elderly using Facebook to “like”a post or post a status. The younger
generation makes full use of the Internet for educational and recreational purposes. How the Internet changes people's lives is
a constant subject of scrutiny. Certainly, educational applications supported by the Internet to foster learning are never out of
the spotlight. Researchers have also endeavored to investigate how to prevent problematic Internet behavior and encourage
sound Internet behavior (Livingstone & Helsper, 2008; Valcke, Bonte, De Wever, & Rots, 2010).
In Taiwan, children begin computer classes in the third grade. Some schools even provide advanced computer classes as
electives. School-aged students are vulnerable to Internet enticements, such as online solicitation or the unprecedented

* Corresponding author.
E-mail addresses: [email protected] (H.-L. Chou), [email protected] (C. Chou), [email protected] (C.-H. Chen).

http://dx.doi.org/10.1016/j.compedu.2015.11.017
0360-1315/© 2015 Elsevier Ltd. All rights reserved.
 H.-L. Chou et al. / Computers & Education 94 (2016) 204e214 205

pleasures of online game playing. Without proper supervision and guidance by parents and teachers, students may access the
Internet without exercising self-discipline. Available research indicates that parents of adolescents are involved in their
children's Internet use, and parents' intervention may be influential (Padilla-Walker, Coyne, Fraser, Dyer, & Yorgason, 2012;
Rosen, Cheever, & Carrier, 2008; Wang, Bianchi, & Raley, 2005). Compared with college students, younger students are more
easily affected by their parents than by their peers. Therefore, how parents act to prevent their children from engaging in
problematic Internet behavior or encourage specific beneficial Internet behavior among their children requires further
investigation.
Certainly, attitudes are related to behavior; however, little research has analyzed the relation between Internet attitudes
and Internet behavior. Therefore, the present study explores the relation between Internet attitudes and Internet behavior in
high-school students and investigates how parenting styles moderate this relation.
Latent class analysis (LCA) was employed to cluster the child-reported parenting styles in Taiwan, and 2nd-order confir-
matory factor analysis was used to validate the 6-T Internet attitude model proposed by Chou, Wu, and Chen (2011, 2013).
Finally, the results derived from the structural regression modeling show the moderating effects of parenting styles on the
relation between Internet attitudes and Internet behavior. A discussion of the implications of the findings is followed by the
statistical analysis.

2. Literature review

2.1. Parenting style

Home is a place in which adolescents spend much time, and parents play an important role in all aspects of their children's
development. Numerous researchers have attempted to determine how parents' actions affect their children's development
using the dyads' behavior, attitudes and values. Related research includes children's attachment relationships (Cohn, Cowan,
Cowan, & Pearson, 1992), children's social anxiety (Greco & Morris, 2002) and children's emotional adjustment (Butler,
Skinner, Gelfand, Berg, & Wiebe, 2007). Researchers have observed particular disciplinary techniques of parents, recorded
typical behavioral responses of children, and measured specific psychological attributes of parents (Baumrind, 1966; Maccoby
& Martin, 1983). Researchers have termed these phenomena parentechild interaction, socialization or parenting styles. These
terms and research contexts were generally unrelated to the Internet.
In the mid-1960s, Baumrind collected and reviewed prior research on parental disciplinary techniques and their related
effects (Baumrind, 1966). She identified seven disciplinary techniques (considered as parental control in most related studies),
including punishment, withdrawal of love, and demand for household responsibilities. Three models of parenting styles were
identified by different combinations of disciplinary techniques - permissive, authoritarian and authoritative. The three
models of parenting styles were confirmed in Baumrind's later research using qualitative observations of parentechild
interaction (Baumrind, 1967).
Maccoby and Martin (1983) categorized Baumrind's disciplinary techniques into two elements: responsiveness and
demandingness. Parental responsiveness refers to how parents respond and adapt to a child's signals, statuses, and needs.
Parental demandingness reflects how parents exert consistent discipline, make maturity demands and encourage inde-
pendent contacts.
Consequently, the permissive parenting style defined by Baumrind was replaced by indulgent and neglecting parenting
styles. Using the two orthogonal dimensions, responsiveness and demandingness, four parenting styles emerged. Authori-
tative parenting refers to both high demandingness and high responsiveness. Indulgent parenting is characterized by low
demandingness but high responsiveness. Neglecting parenting refers to low demandingness and low responsiveness.
Authoritarian parenting refers to high demandingness but low responsiveness (see Fig. 1). Maccoby and Martin's typology of
parenting styles was accepted in Baumrind's later work (Baumrind, 1991). This typology is widely recognized in the literature,
and it is noteworthy that a similar typology of two-dimensional parenting was introduced even earlier with slightly different

Fig. 1. Parenting styles.

206 H.-L. Chou et al. / Computers & Education 94 (2016) 204e214

names, such as warmth/hostility and restrictiveness/permissiveness (Becker, 1964) and warmth/hostility and control/au-
tonomy (Schaefer, 1965).
Recently, researchers have identified the manner in which parents control or respond to children's technology use as
Internet parenting style and have even proposed an additional parenting model - mixed Internet parenting style - to reflect
the median level of parental responsiveness and demandingness regarding children's Internet use (Valcke et al., 2010).
However, most researchers adopt the traditional four types of parenting styles as Internet parenting styles (Chan & Koo, 2011;
Horzum & Bektas, 2014; Ihmeideh & Shawareb, 2014).

2.2. Parenting style and Internet behaviors

Although parents' influence on their children's development has been studied for decades, comparatively little research
has focused on parental control over children's Internet use. Some studies have described the effect of parents' disciplinary
techniques on children's Internet use. Disciplinary techniques represent the various manners in which parents control their
children's Internet use, including active co-viewing and restricting technology. Parental disciplinary techniques are more like
parental control than parental warmth. The related research suggests that applying the disciplinary techniques with tradi-
tional media to children's Internet use could be fruitless (Warren & Bluma, 2002). Parents may not know how to adequately
intervene in children's Internet use (de Morentin, Corte
s, Medrano, & Apodaca, 2014). Meanwhile, researchers have posited
that child-experienced parental disciplinary techniques affect website safeguard effectiveness. More specifically, more robust
parental disciplinary techniques result in less effective website safeguards to prevent children from revealing personal data.
Active disciplinary techniques help reduce personal data disclosure even when website safeguards are unavailable (Lwin,
Stanaland, & Miyazaki, 2008).
Liau, Khoo, and Hwaang (2005) incorporated parental warmth, such as communication, as a predictor to explore the
relation between parents' specific socialization tactics and children's behavior. These authors concluded that specific parental
supervision techniques, particularly the rule of not meeting friends made online, are effective in reducing these incidents,
whereas some other techniques do not work. Children's communication with parents regarding porn-email spam also helped
reduce the occurrence of these incidents.
Due to the affordances of the Internet, Eastin, Greenberg, and Hofschire (2006) extended the scope of disciplinary tech-
niques to general parenting styles, discovering that parents with distinct parenting styles apply different disciplinary tactics
to children's media use. For example, utilizing website block software was the most common mediation technique among
authoritative parents, followed by authoritarian and neglecting parents. Additionally, parenting styles do not affect the
amount of time children spend online, although the location of Internet access affects children's time allocation.
Other researchers concluded that different parenting styles appear to be relatively ineffective in reducing children's risky
behaviors. Parents exhibiting either high control or low control are generally correlated with an increase in children's risky
online behavior. Although parents of high warmth are associated with a decrease in Internet addiction, these parents are also
correlated with more unauthorized acts (Lau & Yuen, 2013).
Conversely, another study revealed that the dimension of parental control in parenting styles could reduce unsafe Internet
use, whereas school-based intervention appears ineffective. Students who experienced parental control are less likely to
reveal personal information on the Internet (Valcke, Schellens, Van Keer, & Gerarts, 2007). Another study stated that
parenting style is strongly related to parents' attitudes and adolescents' Internet behavior on the social networking service
Myspace (Rosen et al., 2008). Parents of older children are less likely to set limits on the use of Myspace and are less concerned
with children's posting of information; thus, these parents often exhibit a neglecting or an indulgent parenting style.
Authoritative parents are more aware of their children's behavior on Myspace, and this type of parenting is therefore more
associated with less risky Myspace behavior in children. However, even authoritative parents are more knowledgeable about
whether their children have revealed personal information on Myspace, the parental awareness that to what extent personal
information is disclosed by their children did not vary by the child's age.
Valcke et al. (2010) identified five Internet parenting styles and observed that authoritative parents composed the ma-
jority. These authors defined the level of Internet use operationally as the frequency and amount of time spent on the Internet.
The researchers also reported that the highest level of child Internet use was associated with a child-perceived permissive
parenting style, whereas the lowest level of use was associated with the child-perceived authoritarian parenting style.
Additionally, parents' Internet behavior and educational background can predict children's level of Internet use.

2.3. T-model of Internet attitude

Several researchers adopted affection, usefulness and perceived control to develop scales to measure computer or Internet
attitudes (Abedalaziz, Jamaluddin, & Leng, 2013; Selwyn, 1997; Tsai, Lin, & Tsai, 2001). Tsai and colleagues proposed a 4-T
model to illustrate Taiwanese students' perception of the Internet and posited perception as fundamental in shaping atti-
tude and behavior (Tsai & Lin, 2004). The 4-T model includes Technology, Tool, Toy, and Tour/Travel. The respondents
perceived the Internet as Technology to improve life. The respondents perceived the Internet as a Tool for information
acquisition and communication and as a Toy for pleasure, particularly in online game playing. They also perceived the Internet
as Tour/Travel to navigate and open up their world. The 4-T model not only measures students’ perception of the positive
effects of the Internet but also identifies nuances among the perceptions.
 H.-L. Chou et al. / Computers & Education 94 (2016) 204e214 207

Chou and colleagues revised Tsai's model to develop a new 6-T model to measure Taiwanese students' Internet attitudes
(Chou et al., 2011, 2013; Chou, Yu, Chen, & Wu, 2009). Sample items of the 6-T questionnaire include “I enjoy using the
Internet” and “It's easy for me to learn new Internet applications.” Tool, Toy, Telephone, Territory, Treasure of Information and
Trade became the six facets of Internet attitudes among Taiwanese students. Students regard the Internet as a Tool to perform
various tasks and as a Toy for entertainment, particularly for online game playing. Telephone indicates that users use the
Internet for communication, and Territory is a concept arising from the era of web 2.0, referring to self-expression on the
Internet. Treasure of information defines the Internet as a huge repository of information, and Trade, the final proposed factor
among the 6-T model, reflects the popularity of buying and selling online. Numerous examples of 6-T research have validated
this Internet attitude model among primary, high school and college students in Taiwan.

2.4. Relation between Internet attitudes and Internet behavior

Attitudes toward the Internet are viewed as an important factor in the acceptance and use of technology (Porter & Donthu,
2006; Teo, Lee, & Chai, 2008); attitudes are thus strongly related to Internet behavior. Notably, Jackson et al. (2003) revealed
that negative Internet attitudes predicted adults’ Internet use better than positive Internet attitudes. When adult users think
that online information is not completely true and that the Internet is less important, they use the Internet more for social
participation. Researchers concluded that less trusting attitudes are more informed attitudes. Experienced Internet users
already understand the concept of Internet risks.
One study used college students as subjects and concluded that college students' positive attitudes toward the Internet are
associated with more frequent use of the Internet for both general and educational purposes (Duggan, Hess, Morgan, Kim, &
Wilson, 2001). Another study similarly found that college students' attitudes toward the Internet are positively correlated
with the students’ Internet use, including frequency, intensity, purposes and variety (Cheung & Huang, 2005). Peng, Tsai, and
Wu (2006) argued that college students who perceived the Internet as Tour or Toy exhibited more positive Internet feeling
and better communicative self-efficacy than students who perceived the Internet as Technology or Tool.
€
Ozcan and Buzlu (2007) used the Online Cognitive Scale to measure problematic Internet use; the scale includes four
dimensions: loneliness/depression, diminished impulse control, distraction and social comfort. These authors observed that
the students who scored higher on the scale were less engaged in online activities related to learning and that the students'
scores on the scale could positively predict their engagement in online activities related to entertainment. Another study also
showed, to some extent, the positive correlation between the Internet attitude and Internet addiction (Masrek, Aziz, & Johare,
2012). Tsai and Lin (2001) observed that high school students’ Internet attitudes could explain some aspects of Internet
addiction, and students who valued the Internet highly inclined toward Internet addiction.

3. Research questions

According to prior research, parents may play a vital role in shaping children's Internet behavior by imparting standards
and by involving themselves in and responding to children's Internet use. Children have varying attitudes toward the Internet,
and Internet attitude has much to do with Internet behavior. Moreover, prior research indicates that certain parenting styles
may be associated with the fluctuation of children's risky Internet behavior, and some research has used parenting style as a
moderator of the effect of parenting practice on the child's development and as a moderator of the relation between values
and related behavior (Darling & Steinberg, 1993; Padilla-Walker, Fraser, & Harper, 2012). The present study investigates how
parenting styles moderate the relation between Taiwanese high school students' Internet attitudes and the students' Internet
behavior. The specific research questions of this study are the following.

(1) Which Internet parenting styles are perceived by Taiwanese high school students?
(2) Can the 6-T Internet attitude model be validated by 2nd-order CFA in our sample?
(3) How do Internet parenting styles moderate the relation between Taiwanese high school students' Internet attitudes
and Internet behavior?

4. Research design

4.1. Instrument

Questionnaires primarily focusing on child-perceived Internet parenting styles, Internet attitudes and specific Internet
behavior were distributed to secondary school students in Taiwan. Four-point Likert scales helped respondents express the
frequency or intensity of agreement with an item and avoid the tendency to report a “neutral” opinion. A higher score
represents a higher level of agreement with the statement.
The questionnaire comprises four sections. The first section collects respondents' demographic data, such as gender, age
and grade level, and their general use of the Internet, including location of use, amount of time spent and purpose. The second
section comprised 13 items related to the students' perceived Internet parenting styles. The respondents answered to what
208 H.-L. Chou et al. / Computers & Education 94 (2016) 204e214

extent they perceived their parents controlled (demandingness) or responded to their Internet use and needs (responsive-
ness). All 13 items were reviewed by two domain experts, ensuring expert validity. The Cronbach's internal consistent reli-
ability coefficient of the parenting style section is approximately .83.
The third section comprised 29 items adopted from the 6-T model developed by Chou et al. (2011) to measure the students'
Internet attitudes. The Cronbach's internal consistent reliability coefficient of this entire section is .94, and the coefficients of
the six factors are approximately .90 with some variation.
The fourth section included 15 items referring to three types of Internet behavior (i.e., online social compensation seeking,
online game addiction, and information security behavior). Every five items are grouped into one type of Internet behavior.
Item 3 is an example of online social compensation seeking: “I'll seek friendship on the Internet if needed.” Item 4 is an
example of online game playing: “If no one limits the time I play an online game, I'll spend a great deal of time on it.” Item 14 is
an example of information security behavior: “I won't use software of questionable origin.” Notably, the first two types of
Internet behavior are problematic, whereas the third type, “information security behavior,” concerns honorable Internet
behavior. The Cronbach's internal consistent reliability coefficients of the three factors are between .72 and .83. Some indices
revealed that the data were appropriate for factor analysis (KMO ¼ .82, Bartlett's Test c2105 ¼ 1293:681, p < .001). The factor
analysis revealed that each set of related 5 items loaded highly on its own factor but lowly on other factors. Thus, we
concluded that the items developed in this section had discriminant validity. For simplicity, we used the factor score as the
measure of a specified Internet behavior in the following analysis.
The statistical analyses were implemented with statistical package PASW 19 and Mplus 7.0.

4.2. Participants and descriptive statistics

Without offering any incentives to the participants, 270 valid questionnaires from high school-aged students were
collected by self-selection. Among the 270 respondents, 128 were male and 152 were female; 151 students were junior-high
school students and 119 were senior-high school students. The students had accessed the Internet for at least four years
because computer classes are introduced at the third grade level. The students had a basic knowledge of computers as well as
the Internet, and they often used the Internet for schoolwork. Parents of high school students may involve themselves in their
children's daily affairs less than the parents of primary school students. However, parenting style remains influential with the
students because most high school students live with their parents. This influence is partially why we consider parenting style
as a context to moderate the relation between children's Internet attitudes and Internet behavior.

4.3. Types of child-perceived parenting styles

Whether the 4-point Likert-scale measure is continuous or ordered-categorical remains controversial. In our research, we
viewed this scale as a categorical measure so that latent class analysis (LCA) would be a better analytic technique than latent
profile analysis (LPA) when performing the clustering analysis. LCA partitions the population into discrete classes based on
the respondents’ response patterns to each variable. We utilized LCA to categorize the perceived parenting styles of Taiwanese
adolescents. Five classes, four classes, three classes and two classes were tested. The resulting indices are listed in Table 1.
The model with the lowest AIC or BIC should be considered the better model (see Table 2). Additionally, for the 3-class
solution, the hypothesis that the 2-class solution is no different from the 3-class solution according to the Vuong-Lo-
Mendell-Rubin Likelihood Test was rejected (p < .001). Consequently, we considered 3 classes to be a favorable number of
types of parenting styles as perceived by adolescents in Taiwan.
In the 3-class classification, there are 81 students in the 1st Class, 126 students in the 2nd Class, and 73 students in the 3rd
Class. We named the 1st Class the authoritative-parenting class because most members in this group achieved high scores on
related items. Item 6 illustrates parental demandingness: “My parents are strict on the rules they set for my Internet use.”
Item 5 is an example of parental responsiveness: “My parents will analyze, explain and communicate with me about how to
use the Internet.”We named the 2nd Class the indulgent-parenting class. Because of the members' low scores on every item,
we concluded that parents in this class care about what their children need in the Internet use, although these parents are less
strict about executing the rules regarding children's Internet use. Item 9 illustrates 2nd Class: “My parents don't care about my
Internet needs.” We named the 3rd Class the neglecting-parenting class. Compared with students in the 2nd Class, students in
the 3rd Class rated higher score on Item 9, which states, “My parents don't care about my needs regarding Internet use.”
Additionally, the parents in the 3rd Class control or respond to children's Internet use less than the parents of the other two
groups do.

Table 1
Related results of LCA.

No. of classes AIC BIC Entropy Vuong-Lo-Mendell-Rubin Likelihood test

2 8666.224 8953.392 .896 <.001
3 8300.121 8732.661 .906 <.001
4 8192.039 8769.971 .913 .7616
5 8133.972 8857.295 .938 .8315
 H.-L. Chou et al. / Computers & Education 94 (2016) 204e214 209

Some facts revealed by the respondents can validate our classification. The students who used the Internet after 10 p.m. on
weekdays represent a relatively small portion of the 1st Class. Students who perceived indulgent and neglecting parenting
were inclined to use the Internet late into the night compared with those who perceived authoritative parenting. Additionally,
there were more high school students than middle school students in the 2nd Class and the 3rd Class (c2ð2Þ ¼ 18:650, p < .001).
Junior high school students, then, perceive their parents as more involved in their Internet use than do senior high school
students. More students in the 1st Class use the Internet for schoolwork-related information (c2ð2Þ ¼ 12:102, p ¼ .002), and
more students in the 3rd Class use the Internet for trade (c2ð2Þ ¼ 11:885, p ¼ .003). Nevertheless, gender does not make a
statistical difference across the three classes (c2ð2Þ ¼ 1:845, p ¼ .398).

4.4. Six-T model of Internet attitudes

Due to controversy over whether the 4-point Likert scale is continuous, we view the scale as ordinal categorical. Unlike the
original research conducted by Chou, Wu, and Chen (2013), we thereby applied the weighted least squares with mean and
variance-adjusted (WLSMV) estimation to run CFA with categorical indicators to validate the 6-T model of Internet attitudes.
The c2ð371Þ value of 1134.634 represents a corrected version of the chi-square statistic in categorical variables. We attributed
the significance of the chi-square test to the large sample size. The remaining model fit indices exhibit an acceptable model
(CFI ¼ .947, RMSEA ¼ .086) (See Table 2).

Table 2
Result of second-order CFA of the 6-T Internet attitudes.

2nd-Order CFA c2 df CFI RMSEA (90% C.I.) WRMR

1134.634 371 .947 .086 [.080, .091] 1.571

Both the first-order and second-order factor loadings are statistically significant. The six factors are listed below:

(1) Tool (4 items): This construct reflects that the respondents value the usefulness of the Internet to perform various tasks.
The respondents agree that the Internet is a good helper and also an indispensable tool in daily life.
(2) Telephone (7 items): This construct indicates that the respondents use the Internet to communicate with others and
make friends. A sample item is the statement that the Internet is a good place for friends to exchange opinions.
(3) Territory (3 items): This construct indicates that the respondents love to express themselves online. People use either a
blog to share relevant content or Wiki to collaboratively create content. Additionally, people either record every bit of
their lives or post comments in a particular online community.
(4) Toy (6 items): In this construct, the respondents regard the Internet as an aid to entertainment. Item 8, for example,
states that the Internet is a place “where I can spend my leisure time.” The items that compose this factor also reflect a
certain level of affection, including statements indicating that using the Internet can ease stress or prevent boredom.
(5) Treasure of information (7 items): This construct defines the Internet as a borderless ocean holding a huge amount of
information. Item 2 is a good example of this factor: “I can acquire useful information through the Internet.” Notably,
Item 7 loads comparatively lower on this factor. The factor loading is .25 but statistically significant. Item 7 reads, “The
Internet enables me to access harmful information, such as pornography and violence.” This item was self-developed to
emphasize that users gradually grasp that there are many types of information online, correct or incorrect, useful or
harmful.
(6) Trade (2 items): This construct refers to the notion of online shopping. Item 14 exemplified this factor: “I use the
Internet to buy and sell things.”

High rating in this section indicates that the respondent values the Internet highly. Distinct factor loadings on each 1st-
order construct reveal that the respondent perceives the functions of the Internet at different levels. Overall, the respondents
value the Internet as a Tool the most and as Trade the least (see Fig. 2).

5. How parenting styles moderate the relation between attitudes and behavior

We incorporated the Internet attitude measurement model and the path analysis of the relation between Internet atti-
tudes and behavior as the structural regression model (SR Model). For simplicity, the 1st-order factor scores derived from the
above 6-T model were adopted instead of using the 29 indicators. Factor scores of specific Internet behavior were also
adopted.
To test the group difference as a function of parenting style, multi-group analysis was employed. The significance of
regression paths across different parenting classes was tested by the Wald chi-square test. Table 3 indicates that the structural
regression models differ across the three groups.
210 H.-L. Chou et al. / Computers & Education 94 (2016) 204e214
Fig. 2. Six-T Internet attitudes.
 H.-L. Chou et al. / Computers & Education 94 (2016) 204e214 211

Table 3
Chi-square difference test of the Structural Regression Model.

SR Model Chi-square df CFI RMSEA (90% C.I.) SRMR

Full Model 190.083 92 .957 .107 [0.085, .129] .043
Authoritative 52.805
Indulgent 82.904
Neglecting 54.375
Constraint Model 210.935 98 .949 .111 [.090,.132] .09
Authoritative 59.116
Indulgent 93.673
Neglecting 58.146

Dc2 ¼ 19:794 Ddf ¼ 6 p ¼ :003.

The results indicate that Internet attitudes can explain 20.1%, 23.8% and 1.1% of the variance in Internet behavior in the
authoritative parenting class. Internet behavior includes “online game addiction”, “online social compensation seeking” and
“information security behavior.” Internet attitudes can explain 13.2%, 8.7% and 15.3% of the variance in Internet behaviors in
the indulgent parenting class. Internet attitudes can explain 7.6%, 14.5% and 1.1% of the variance in Internet behavior in the
neglecting-parenting class.
Internet attitudes are statistically significant in predicting online social compensation and game addiction across the three
classes, whereas attitudes are only statistically significant in predicting information security behavior in the indulgent-
parenting class.
The disparity in the standardized regression coefficients on the path leading to information security behavior between the
authoritative and indulgent parenting classes when constraining the other paths equally across the groups is statistically
significant according to the Wald test (Dc2 ¼ 11:085; Ddf ¼ 1; p < :001). The disparity in standardized regression coefficients
on the path leading to information security behavior between the neglecting- and indulgent-parenting classes when con-
straining the other paths equally across the groups is statistically significant according to the Wald test
(Dc2 ¼ 9:339; Ddf ¼ 1; p ¼ :002). That is, students in the indulgent-parenting class who value the Internet more highly are
more likely to exhibit secure Internet behavior than students in the authoritative- and neglecting-parenting classes (see
Fig. 3).

6. Discussion and conclusion

This study identified three perceived parenting styles among high school-aged students in Taiwan. The reason that the
authoritarian parenting style is not included may be because of the cultural context. In Chinese society, children are inured to
parental training and involvement in both academic and non-academic domains. Thus, Chinese children barely report the
authoritarian parenting style, while Western adolescents define this style as demanding. Chao (1994) addressed the concept
of guan (synonymous with “training” in English) to contend that the authoritative and authoritarian parenting styles are
somewhat indistinguishable in Chinese society because Asians’ understanding of training involves demandingness and
responsiveness. Our parenting clusters are generally consistent with this viewpoint. The authoritative parenting in our
classification can be merged with the authoritarian parenting style.
The 6-T Internet attitude model is again validated by the 2nd-order CFA in this sample. We developed and added Item 7 to
emphasize that the Internet inherently retains various pieces of information, correct or incorrect. Certainly, online infor-
mation may be harmful if used improperly. Although Item 7 loads lowly on the construct of “Treasure of information”, it is
statistically significant. Item 7 reflects that students gradually experience the diversity of information revealed online.
Although the students perceive the Internet as a “Treasure of information”, they must always take information verification
into account. This caveat should direct the design of related programs to educate students. We hope that future research on
the Internet attitude scale will take this item into consideration.
Students who value the Internet highly have some tendency to seek social compensation. Parenting styles do not exert any
moderating effect on the behavior of online social compensation seeking. That is, those who value the Internet highly may
seek online social compensation regardless of perceived parenting style. This conclusion suggests parents should not merely
rely on certain parenting techniques to safeguard appropriate online interpersonal relationships for their children.
Similarly, students who value the Internet highly may have a tendency toward online game addiction. We suggest that
parents respond to children's needs and monitor their children's use of the Internet simultaneously. Parents may not be able
to successfully supervise or control children's online game playing; however, satisfying or neglecting children's needs could
be detrimental to children's Internet use, strengthening, for example, their addiction to online game playing.
The study results show the moderating effect of different parenting styles on the relation between children's Internet
attitudes and their information-security behavior. To our surprise, Internet attitudes can only predict sound information-
security behavior in the indulgent-parenting class. One reason may be that the wording of items regarding information
security behavior in our survey includes whether students use software of questionable origin. Parents of the indulgent-
parenting style could satisfy children's needs regarding Internet use; hence, students in this class have an excellent chance
of purchasing software they need. The non-significant path coefficients revealed in the other two classes convey the message
 212
H.-L. Chou et al. / Computers & Education 94 (2016) 204e214
Fig. 3. Structural regression model of different parenting styles.
 H.-L. Chou et al. / Computers & Education 94 (2016) 204e214 213

that students are not adequately acquainted with the necessary knowledge and skills to use the Internet safely. We
recommend that parents guide children in related matters, that teachers design related programs, and that policy makers
enforce related rules.
This present study observed that highly valued Internet attitudes contribute to problematic Internet behavior and that
highly valued Internet attitudes also lead to safe Internet behavior. We must be cautious in enhancing students’ Internet
attitudes. We identified the moderation effect of parenting styles on the relation between Internet attitudes and Internet
behavior. A detailed understanding of how parenting styles moderate this relation certainly lends itself to preventing risks
children might encounter online. This knowledge also helps stimulate the proper relationships that parents and children
should cultivate. Because this exploratory research was conducted in Taiwan and information was collected by self-selection,
generalization remains limited because of the cultural context and the sampling bias. The moderation effect of parenting
styles requires further investigation.
Researchers interested in this topic can conduct similar studies using different statistical methods and different scales to
measure Internet attitudes or different clustering techniques to group the parenting styles. Because the parenting styles are
multi-faceted and complicated, we also suggest a qualitative analysis on differentiating types of perceived parenting styles.
Additionally, researchers can investigate how parenting styles moderate the relation between Internet attitudes and other
types of children's Internet behavior patterns.

Acknowledgments

This study was supported by the Ministry of Science and Technology in Taiwan under grant number MOST 100-2511-S-
009-002-MY3. The authors would like to thank Dr. Yih-Lan Liu for her thoughtful input in designing the questionnaires. We
are also indebted to Dr. Jiun-Yu, Wu for his comments and instructions on the statistical analysis.

References

Abedalaziz, N., Jamaluddin, S., & Leng, C. H. (2013). Measuring attitudes toward computer and internet usage among postgraduate students in Malaysia.
Turkish Online Journal of Educational Technology-TOJET, 12(2), 200e216.
Baumrind, D. (1966). Effects of authoritative parental control on child behavior. Child Development, 887e907. http://dx.doi.org/10.2307/1126611.
Baumrind, D. (1967). Child care practices anteceding three patterns of preschool behavior. Genetic Psychology Monographs, 75(1), 43e88.
Baumrind, D. (1991). The influence of parenting style on adolescent competence and substance use. The Journal of Early Adolescence, 11(1), 56e95. http://dx.
doi.org/10.1177/0272431691111004.
Becker, W. C. (1964). Consequences of different kinds of parental discipline. In M. L. Hoffman, & L. W. Hoffman (Eds.), Review of Child Development Research
(vol. 1, pp. 169e208). New York: Russell Sage Foundation.
Butler, J. M., Skinner, M., Gelfand, D., Berg, C. A., & Wiebe, D. J. (2007). Maternal parenting style and adjustment in adolescents with type I diabetes. Journal of
Pediatric Psychology, 32(10), 1227e1237. http://dx.doi.org/10.1093/jpepsy/jsm065.
Chan, T. W., & Koo, A. (2011). Parenting style and youth outcomes in the UK. European Sociological Review, 27(3), 385e399. http://dx.doi.org/10.1093/esr/
jcq013.
Chao, R. K. (1994). Beyond parental control and authoritarian parenting style: understanding Chinese parenting through the cultural notion of training.
Child Development, 65(4), 1111e1119. http://dx.doi.org/10.1111/1467-8624.ep7252822.
Cheung, W., & Huang, W. (2005). Proposing a framework to assess Internet usage in university education: an empirical investigation from a student's
perspective. British Journal of Educational Technology, 36(2), 237e253.
Chou, C., Wu, H.-C., & Chen, C.-H. (2011). Re-visiting college students' attitudes toward the internet-based on a 6-T model: gender and grade level difference.
Computers & Education, 56(4), 939e947. http://dx.doi.org/10.1016/j.compedu.2010.11.004.
Chou, C., Wu, H.-C., & Chen, C.-H. (2013). Tool, toy, telephone, territory, trade, or treasure of information: a cross-sectional study of Taiwanese students'
attitudes toward the Internet. Chinese Journal of Communication, 6(2), 202e220. http://dx.doi.org/10.1080/17544750.2013.785671.
Chou, C., Yu, S.-c., Chen, C.-h., & Wu, H.-C. (2009). Tool, toy, telephone, territory, or treasure of information: elementary school students' attitudes toward
the Internet. Computers & Education, 53(2), 308e316. http://dx.doi.org/10.1016/j.compedu.2009.02.003.
Cohn, D. A., Cowan, P. A., Cowan, C. P., & Pearson, J. (1992). Mothers' and fathers' working models of childhood attachment relationships, parenting styles,
and child behavior. Development and Psychopathology, 4(03), 417e431. http://dx.doi.org/10.1017/S095457940000087.
Darling, N., & Steinberg, L. (1993). Parenting style as context: an integrative model. Psychological bulletin, 113(3), 487.
Duggan, A., Hess, B., Morgan, D., Kim, S., & Wilson, K. (2001). Measuring students' attitudes toward educational use of the internet. Journal of Educational
Computing Research, 25(3), 267e282.
Eastin, M. S., Greenberg, B. S., & Hofschire, L. (2006). Parenting the internet. Journal of Communication, 56(3), 486e504. http://dx.doi.org/10.1111/j.1460-
2466.2006.00297.x.
Greco, L. A., & Morris, T. L. (2002). Paternal child-rearing style and child social anxiety: investigation of child perceptions and actual father behavior. Journal
of Psychopathology and Behavioral Assessment, 24(4), 259e267. http://dx.doi.org/10.1111/jomf.12107.
Horzum, M. B., & Bektas, M. (2014). Examining the Internet use aim and internet parental style of primary school students in terms of various variables.
Croatian Journal of Education, 16(3), 745e778.
Ihmeideh, F. M., & Shawareb, A. A. (2014). The association between internet parenting styles and children's use of the internet at home. Journal of Research in
Childhood Education, 28(4), 411e425. http://dx.doi.org/10.1080/02568543.2014.944723.
Jackson, L. A., Von Eye, A., Barbatsis, G., Biocca, F., Zhao, Y., & Fitzgerald, H. E. (2003). Internet attitudes and internet use: some surprising findings from the
HomeNetToo project. International Journal of Human-Computer Studies, 59(3), 355e382. http://dx.doi.org/10.1016/S1071-5819(03)00069-7.
Lau, W. W., & Yuen, A. H. (2013). Adolescents' risky online behaviours: the influence of gender, religion, and parenting style. Computers in Human Behavior,
29(6), 2690e2696. http://dx.doi.org/10.1016/j.chb.2013.07.005.
Liau, A. K., Khoo, A., & Hwaang, P. (2005). Factors influencing adolescents engagement in risky Internet behavior. CyberPsychology & Behavior, 8(6), 513e520.
http://dx.doi.org/10.1089/cpb.2005.8.513.
Livingstone, S., & Helsper, E. J. (2008). Parental mediation of children's internet use. Journal of broadcasting & electronic media, 52(4), 581e599. http://dx.doi.
org/10.1080/08838150802437396.
Lwin, M. O., Stanaland, A. J., & Miyazaki, A. D. (2008). Protecting children's privacy online: how parental mediation strategies affect website safeguard
effectiveness. Journal of Retailing, 84(2), 205e217. http://dx.doi.org/10.1016/j.jretai.2008.04.004.
Maccoby, E. E., & Martin, J. A. (1983). Socialization in the context of the family: parent-child interaction. In P. H. Mussen (Ed.), Handbook of child psychology:
Formerly Carmichael's manual of child psychology.
214 H.-L. Chou et al. / Computers & Education 94 (2016) 204e214

Masrek, M. N., Aziz, N. S. A., & Johare, R. (2012). The relationship between internet attitude and internet addiction. Asian Journal of Information Technology,
11(4), 125e130. http://dx.doi.org/10.3923/ajit.2012.125.130.
de Morentin, J. I. M., Corte
s, A., Medrano, C., & Apodaca, P. (2014). Internet use and parental mediation: a cross-cultural study. Computers & Education, 70,
212e221. http://dx.doi.org/10.1016/j.compedu.2013.07.036.
€
Ozcan, N. K., & Buzlu, S. (2007). Internet use and its relation with the psychosocial situation for a sample of university students. CyberPsychology & Behavior,
10(6), 767e772. http://dx.doi.org/10.1037/a0027720.
Padilla-Walker, L. M., Coyne, S. M., Fraser, A. M., Dyer, W. J., & Yorgason, J. B. (2012a). Parents and adolescents growing up in the digital age: latent growth
curve analysis of proactive media monitoring. Journal of adolescence, 35(5), 1153e1165. http://dx.doi.org/10.1016/j.adolescence.2012.03.005.
Padilla-Walker, L. M., Fraser, A. M., & Harper, J. M. (2012b). Walking the walk: the moderating role of proactive parenting on adolescents' value-congruent
behaviors. Journal of adolescence, 35(5), 1141e1152. http://dx.doi.org/10.1016/j.adolescence.2012.03.003.
Peng, H., Tsai, C. C., & Wu, Y. T. (2006). University students' self-efficacy and their attitudes toward the Internet: the role of students' perceptions of the
Internet. Educational Studies, 32(1), 73e86. http://dx.doi.org/10.1080/03055690500416025.
Porter, C. E., & Donthu, N. (2006). Using the technology acceptance model to explain how attitudes determine Internet usage: the role of perceived access
barriers and demographics. Journal of business research, 59(9), 999e1007. http://dx.doi.org/10.1016/j.jbusres.2006.06.003.
Rosen, L. D., Cheever, N. A., & Carrier, L. M. (2008). The association of parenting style and child age with parental limit setting and adolescent MySpace
behavior. Journal of Applied Developmental Psychology, 29(6), 459e471. http://dx.doi.org/10.1016/j.appdev.2008.07.005.
Schaefer, E. S. (1965). Children's reports of parental behavior: an inventory. Child development, 36(2), 413e424. http://dx.doi.org/10.2307/1126465.
Selwyn, N. (1997). Students' attitudes toward computers: validation of a computer attitude scale for 16e19 education. Computers & Education, 28(1), 35e41.
http://dx.doi.org/10.1016/S0360-1315(96)00035-8.
Teo, T., Lee, C. B., & Chai, C. S. (2008). Understanding pre-service teachers' computer attitudes: applying and extending the technology acceptance model.
Journal of computer assisted learning, 24(2), 128e143. http://dx.doi.org/10.1111/j.1365-2729.2007.00247.x.
Tsai, C.-C., & Lin, S. S. (2001). Analysis of attitudes toward computer networks and Internet addiction of Taiwanese adolescents. CyberPsychology & Behavior,
4(3), 373e376.
Tsai, C.-C., & Lin, C.-C. (2004). Taiwanese adolescents' perceptions and attitudes regarding the internet: Exploring gender differences. Adolescence, 39(156),
725e734.
Tsai, C.-C., Lin, S. S., & Tsai, M.-J. (2001). Developing an Internet attitude scale for high school students. Computers & Education, 37(1), 41e51.
Valcke, M., Bonte, S., De Wever, B., & Rots, I. (2010). Internet parenting styles and the impact on Internet use of primary school children. Computers &
Education, 55(2), 454e464. http://dx.doi.org/10.1016/j.compedu.2010.02.009.
Valcke, M., Schellens, T., Van Keer, H., & Gerarts, M. (2007). Primary school children's safe and unsafe use of the Internet at home and at school: an
exploratory study. Computers in Human Behavior, 23(6), 2838e2850. http://dx.doi.org/10.1016/j.chb.2006.05.008.
Wang, R., Bianchi, S. M., & Raley, S. B. (2005). Teenagers' internet use and family rules: a research note. Journal of Marriage and Family, 67(5), 1249e1258.
http://dx.doi.org/10.1111/j.1741-3737.2005.00214.x.
Warren, R., & Bluma, A. (2002). Parental mediation of children's Internet use: the influence of established media. Communication Research Reports, 19(1),
8e17. http://dx.doi.org/10.1080/08824090209384827.