ASM Sc. J.

, 13, Special Issue 5, 2020 for APRU2018, 83-95

Developing and Testing the Psychometric

Properties of the Parental Digital Security (P-
Dis) Questionnaire for Malaysian Parents
Muhd Zulfadli Hafiz Ismail1,2∗, Nik Daliana Nik Farid1,3 and Rafdzah Ahmad Zaki1,3

1University of Malaya (UM). Department of Social and Preventive Medicine, Faculty of

Medicine, University of Malaya, Kuala Lumpur, Malaysia
2
Ministry of Health Malaysia (MOH), Wilayah Persekutuan Putrajaya, Malaysia
3
University of Malaya Medical Centre (UMMC). Department of Public Health, University of
Malaya Medical Centre, Kuala Lumpur, Malaysia

Parental digital security is an important aspect of keeping children safe online and of curbing poor
online behaviours among children. The existence of validated tools to measure parental digital
security is crucial to improve parents’ digital-security practices in the community. As a type of
protective behaviour, parental digital security can be explored based on e stablished protective
behavioural frameworks, including Protection Motivation Theory (PMT). The objective of this study
is to develop and test the psychometric properties of a novel parental digital -security (P-Dis)
questionnaire based on PMT among Malaysian parents. This study involved parents whom children
were both internet users. It used a method of developing an instrument that comprised three stages:
item development, scale development and psychometric-properties testing. A 51-item questionnaire
was produced that covered 9 factors reflecting the PMT domains. The Cronbach’s alpha coefficient
of these factors ranged from 0.79‒0.94. The factor analysis also showed that the 9 factors accounted
for 73.8% of the total variance. These findings indicated that the parental digital-security
questionnaire developed fulfilled its psychometric properties and is suitable for use by Malaysian
parents. Further research is needed to validate the questionnaire for other populations.
Keywords: parental digital security; questionnaire; validation; Protection Motivation Theory

I. INTRODUCTION Terms such as ‘internet safety’ and ‘electronic safety’ have

emerged in the literature, reflecting self-protection, which
The present study focused on parental digital security. In the combines both the technical and behavioural aspects of
literature, the concept of digital security varied across security (Lorenz, 2017). Thus, to reflect self-protection in
studies. Generally, as proposed by Lorenz (2017), digital both its technical and behavioural aspects, digital security
security is part of three inter-related concepts: information can be defined as the ability to maintain security and safety
security, computer security and digital safety itself. Both online.
information security and computer security focus on the Therefore, parental digital security refers to parents’
technical measures taken to protect oneself from threats that practices aimed at keeping their children safe online, and it
include data corruption, breach of confidentiality and comprises both technical and behavioural components.
property theft (Lorenz, 2017). In contrast, digital safety Parents play a major role in empowering their children to
focuses on human interaction and behaviour when dealing engage in responsible online use (Cyber Security Malaysia,
with people and information online (Lorenz, 2017). Digital 2014) and to curb poor online behaviours, both of which help
safety refers to internet crimes, including identity theft, to reduce cyber issues among children and young
stalking, cyberbullying and privacy breaches (Lorenz, 2017).

*Corresponding author’s e-mail: Zulfadlihafiz1@gmail.com

 ASM Science Journal, Volume 13, Special Issue 5, 2020 for APRU2018

adolescents. Parents can do this through appropriate and response costs related to a particular protective
parental digital-security practices. The existence of validated behaviour (Maddux and Rogers, 1983). Threat appraisal is
tools with which to measure parental digital security is based on susceptibility to risks, perceived vulnerabilities and
essential to assessing and understanding parents’ needs and maladaptive rewards related to not performing the
subsequently improving their digital-security practices protective behaviour (Maddux and Rogers, 1983). Although
regarding their children. generally PMT has been used to explain one’s own protection
However, very limited instruments are available for motivation, it has also been used to explain the motivation to
measuring parental digital-security practices, particularly in protect others. For instance, Nathanson (2001) applied PMT
the Malaysian context. The tools available were designed to understand parental mediation of children’s use of sexual
mainly in Europe and North America. In Asia, the majority and violent television, discovering that PMT was able to
of studies on digital security were found to originate in explain such protective behaviour by parents. Hence,
China, South Korea and Japan. The focus areas of these exploring parental digital-security practices based on PMT
studies also varied. Studies of parents’ online mediation can be valuable. Therefore, the objective of the present study
techniques varied in terms of the children’s age groups and was to develop and test the psychometric properties of a
the types of online behaviour included (Hwang and Jeong, parental digital-security (P-Dis) questionnaire based on
2015; Sonck et al., 2013; Nikken and Jansz, 2014). For PMT among Malaysian parents.
instance, Sonck et al. (2013) explored online mediation
technique in parents who had children aged 9‒16 years in the II. MATERIALS AND
Dutch context and excluded preschool children. Hwang et al. METHOD
(2017) focused on mediation techniques related to
smartphone use among children in Korea. Nikken et al.
The target population of this scale-development study was
(2014) developed scales for online mediation techniques for
Malaysian parents whom children were younger than 18
children aged 2‒12 years in Holland. None of these
years old and were all internet users. Three major stages were
researchers’ tools specifically addressed parents’ cognitive
involved in developing and testing the questionnaire’s
processes when adopting parental digital-security practices.
psychometric properties: item development, scale
Exploring these cognitive processes is crucial to
understanding parents and effecting behavioural changes development and psychometric properties evaluation. Item

(Willingham, 2007). Thus, there is a clear gap regarding the development involved item generation, content validation

existence of a tool to assess parental digital-security and translation. Scale development involved cognitive
practices that is culturally acceptable in the Malaysian debriefing and test-retest reliability. Testing the
context, covers general online activities, is suitable for psychometric properties involved testing the factor analysis
children and adolescents at all stages and examines parents’ and internal consistency of the questionnaire’s final version.
cognitive processes. Such a tool is essential, as the cyber- Figure 1 depicts these stages.
parenting field is not well understood and researched in the
Malaysian setting from parents’ perspectives (UNICEF
Malaysia, 2014).
As a type of protective behaviour, parental digital security
can be explored based on established protective behavioural
frameworks, including the Protection Motivation Theory
(PMT), which was developed by Rogers in 1975 and revised
by Maddux and Rogers (1983). PMT is a cognitive-based
theory that explains individual protective behaviour
(Maddux and Rogers, 1983). Central to the theory are two
cognitive processes that influence a person’s intention to
adopt a particular protective behaviour: coping appraisal
and threat appraisal (Maddux and Rogers, 1983). Coping Figure 1. Questionnaire-development stages in the present
study
appraisals are determined by response efficacy, self-efficacy

84
 ASM Science Journal, Volume 13, Special Issue 5, 2020 for APRU2018

A. Item-development Stage Table 1. Initial Constructs and Their Definitions

In item development, the initial step, item generation,

involved two types of approaches: the deductive and the
inductive (Hinkin, 1995). The deductive approach was used
to generate items through a review of the literature on
existing questionnaires. The inductive approach was used to
engage with parents through online surveys and with experts
through discussions. An online survey, which used Google
Survey, asked three open-ended questions: ‘What are your
concerns when your children are online?’, ‘What actions have
you taken to ensure the safety of your children while they are
online?’ and ‘What are the barriers to taking actions to keep
your children safe online?’. Then, the themes that emerged
from this online survey, the inventory of items from the
literature review and the domains obtained from reviewing
the literature were discussed by a group of experts, which
comprised the researcher, stakeholders from Cybersecurity
Malaysia, public-health specialists, an expert on cyber-
After this content validation, the questionnaire items were
parenting, parents’ representatives and an expert on
translated into the Malay language, using the forward-
adolescent health. Based on the feedback from this group,
backward technique. The procedures were adapted from the
thematic analysis was conducted on the items initially
guidelines by Gullemin and Beaton (Guillemin et al., 1993;
generated. A preliminary set of items was produced that
Beaton et al., 2000). The initial step involved two
reflected the initial constructs intended to be measured.
independent translators proficient in both English and Malay
Table 1 contains this initial set of constructs and their
but whose native language was Malay. They translated the
definitions.
questionnaire from English to Malay. Then, based on the
Next, the items produced underwent content validation by
same guidelines (Beaton et al., 2000; Guillemin et al., 1993),
experts from the fields of interest, including digital
the two versions were compared and synthesised by a
citizenship, cybersecurity, public health, child and adolescent
committee comprising the researcher, stakeholders from
health, cyber-parenting and anthropology. These experts
CyberSecurity Malaysia and a cyber-parenting expert. A
examined each item using a four-scale scale that measured its
single translated version of the questionnaire was produced.
clarity and relevance to the construct to be measured
Next, the reverse translation was conducted: The Malay
(Streiner and Norman, 2008). The level of agreement among
version was translated into English by two other translators
these experts on content validity in terms of clarity and
working independently, both of whom were proficient in both
relevance was analysed using the Content Validity Index
languages. Similar to the process of forward translation, in
(CVI), with the aim of obtaining a value of individual CVI (I-
the back-translation, the same committee compared the two
CVI) that was at least 0.8 and a value for overall CVI that was
versions with the original English version. A consensus was
at least 0.9 (Streiner and Norman, 2008; Polit and Beck,
achieved, and the final version in both languages was
2006; Lynn, 1986). Two rounds of content validation were
completed.
conducted by experts. The first round included six experts,
and feedback from this round was analysed and revisions
based on the analysis were incorporated. Then, the second
round of content validation involved four experts.

85
 ASM Science Journal, Volume 13, Special Issue 5, 2020 for APRU2018

B. Scale-development Stage (UMREC) using reference number UM.TNC2/UMREC–211.

Participation in the study was voluntary, consent was

In scale development, the mental processes and respondent obtained prior to data collection and the respondents’

burden were assessed by cognitive debriefing procedure, confidentiality was maintained throughout the study.

involving 10 participants who were recruited using purposive

sampling based on their age, ethnicity, gender and education III. RESULT
level. This cognitive debriefing used the verbal probe
technique (Di Lorio, 2005). Participants were first asked to A. Item Development
answer the questionnaire, and then the debriefing session
was conducted to obtain their feedback. During the item-generation step in the item-development
Next, a test-retest reliability assessment was conducted to stage, 69 parents responded online through the Google
assess the consistency of the measurements regarding each of Survey. Table 2 shows their sociodemographic characteristics.
the 36 participants when tested at different times (Vitoratou From the parents’ responses, the literature reviewed and the
S et al., 2009). This test-retest reliability assessment involved discussion with experts, 52 items were generated and mapped
participants answering the questionnaire again after two according to the PMT framework. The items regarding
weeks (Tafforeau et al., 2005). The test-retest reliability was perceived susceptibility and perceived severity (Sections B
analysed based on the intra-class correlation coefficient (ICC) and C, respectively) were derived from the responses on
(Koo and Li, 2016). A target of 0.5 and higher was deemed as concerns about online threats. The items for perceived self-
adequate test-retest reliability (Koo and Li, 2016). efficacy (Section D), perceived response efficacy (Section E)
and digital-security practices (Section H) were derived from
C. Psychometric-properties Evaluation the responses about digital-security practices. The literature
review also contributed to the generation of items,

The scale’s psychometric properties were tested by particularly about digital-security practices (Section H). The

determining the factors through Exploratory Factor Analysis items for response cost (Section F) and maladaptive rewards

(EFA) and internal consistency of the questionnaire’s final (Section G) were derived from the responses about barriers to

version. This test was conducted using 316 parents. The digital-security practices.

recruitment process collected the parents’ demographic In the subsequent two-round content-validation process,

profiles and the self-reported measures that they use. To all items for the respective sections were deemed relevant by

produce the final version of the questionnaire, factors were all the experts in the first round, meaning that individual

extracted using principal axis factoring and the Promax items had CVIs > 0.8 and that each section had an overall CVI >

rotation method (Worthington and Whittaker, 2006). Items 0.9. However, the clarity scores for Sections B and C did not

with a factor load of < 0.4 or having cross-loading issues were meet the threshold. Four items (B1, B2, B4 and B7) had I-

removed and re-analysed (Hair et al., 2010). The reliability of CVIs < 0.8 (0.67).

the final version of the questionnaire was measured using The overall CVI for Section B was 0.81. In Section C, three

inter-item correlation, corrected item-total correlation and items (C2, C4 and C7) had I-CVIs < 0.8 (0.67). The overall

Cronbach’s alpha. Values > 0.3 for the corrected item-total CVI for Section C was 0.83. After the first round of content

correlation (Cristobal et al., 2007) and > 0.7 for the validation, two additional items were created: B8 and C8.

Cronbach’s alpha were acceptable (Cortina, 1993). After additional adjustments were made based on the first

This study was registered in the National Medical Research round, the second round of content validation found I-CVIs

Register (NMRR) as number NMRR-17-3093-39434 (IIR). of 1 and overall CVIs of 1 for all domains regarding clarity

Ethical clearance was obtained from the Medical Research (Table 3). By the end of the content validation process, 54

and Ethics Committee (MREC), Ministry of Health Malaysia items had been generated. Table 4 contains the constructs,

and the University Malaya Research Ethics Committee the item numbers and the actual items.

86
 ASM Science Journal, Volume 13, Special Issue 5, 2020 for APRU2018

Table 2. Sociodemographic Characteristics of Online Survey

Respondents

Table 3. Number of Items and Overall CVIs for Constructs

throughout Content Validity Process

87
 ASM Science Journal, Volume 13, Special Issue 5, 2020 for APRU2018

Table 4. Final Questionnaire after Content Validation Process

Construct Item Item

no.
Perceived Be bullied (harassed, threatened and/or intimidated) online.
B1
susceptibility
(When your child Be spending more time online than he/she should.
is using the B2
internet, in your Be exposed to adult content (e.g. pornography, violence, gambling).
opinion, how likely B3
or unlikely is it Have his/her personal information obtained without his/her knowledge or consent.
that he/she will) B4
Be approached online by a person he/she does not know.
B5
Exchange sexual messages and/or images with other people online.
B6
Be exposed to online content promoting self-harm (e.g. websites that encourage suicide, eating disorders, drug
B7 use).

Be exposed to online content that promotes hate, extreme views or terrorism.

B8
Perceived severity A child is being bullied (harassed, threatened and/or intimidated) online.
C1
(In your opinion,
how serious are A child is spending more time online than he/she should.
these issues to C2
you?)
A child is exposed to adult content (including pornography, violence, gambling).
C3
A child’s personal information is obtained without his/her knowledge or consent.
C4
A child is approached online by a person he/she does not know.
C5
A child exchanges sexual messages and/or images with other people online.
C6
A child is exposed to online content that promotes self-harm (e.g. websites that encourage suicide, eating
disorders, drug use).
C7

A child is exposed to online content that promotes hate, extreme views or terrorism.
C8
Perceived self- I am confident in discussing and giving advice to my child about online safety.
D1
efficacy (How
much do you agree I am confident in my knowledge about keeping my child safe online.
or disagree with D2
the following
I am comfortable with using the internet together with my child.
statement?) D3
I am confident in imposing rules on internet use on my child.
D4
I am confident in using filtering and monitoring software (parental control applications).
D5
I am comfortable with restricting my child to using the internet only when I am around.
D6
I am confident in checking my child’s online activities after my child has been online.
D7
Perceived response Discussing online safety with my child will keep him/her safe online.
E1
efficacy (How
much do you agree Having the appropriate knowledge will keep my child safe online.
or disagree with E2
the following
Using the internet together with my child will keep him/her safe online.
statement?) E3
Imposing internet rules on my child will keep him/her safe online.
E4
Using filtering and monitoring software (parental control applications) will keep him/her safe online.
E5
Restricting my child to using the internet only when I am around will keep him/her safe online.
E6
Checking my child’s online activities after he/she has used the internet will keep him/her safe online.
E7
Perceived response F1 Discussing online safety with my child is troublesome for me.
cost (How much
do you agree or F2 It takes a lot of effort to acquire appropriate knowledge about online safety.

88
 ASM Science Journal, Volume 13, Special Issue 5, 2020 for APRU2018

disagree with the

following F3 It takes a lot of effort to use the internet together with my child.
statement?)
F4 Ensuring that my child follows internet rules is troublesome for me.

F5 Ensuring that filtering and monitoring software (parental control applications) are working can be troublesome
for me.

F6 Restricting my child to using the internet only when I am around requires a lot of effort.

F7 Checking my child’s online activities after he/she has been online requires a lot of effort.

Perceived G1 Not discussing online safety with my child will help to make him/her more independent.
maladaptive
reward (How much Allowing my child to use the internet on his/her own will enable me to focus on my own interests.
G2
do you agree or
disagree with the G3 Not imposing internet rules on my child will make him/her happy.
following
By not putting up filtering and monitoring software (parental control applications), my child can use the
statement?)
G4 internet freely.

By not checking my child’s online activities after he/she is online, I am respecting his/her rights.
G5
Parental digital- Discuss online safety with your child.
H1
security practice
(How often do Have conversations with your child about how to handle unknown people online.
you) H2
Discuss with your child how to protect personal information online.
H3
Have conversations on what to do if he/she is bullied or harassed online.
H4
Use the internet together with your child.
H5
Tell your child when/how long to use the internet.
H6
Tell your child which websites/social networks he/she can visit.
H7
Tell your child what he/she can and cannot do online.
H8
Ensure that filtering and monitoring software (parental control applications) are present.
H9
Restrict your child to using the internet only when you are present.
H10
Check the websites that your child has visited.
H11
Check which friends or contacts your child has added to a social networking profile.
H12

B. Scale Development deleted, leaving 52 items remaining to be further tested for

their psychometric properties.

The cognitive debriefing of the 10 participants found that

most items were well understood and suitable. However, C. Psychometric Properties Test
items F2 and F3 required increased clarity, which was
achieved by defining the term ‘effort’. Item H12 was deleted, Table 5 shows the sociodemographic characteristics of the
as it was deemed irrelevant to some parents. 316 respondents involved in factor analysis and internal
The remaining 53 items underwent test-retest process consistency testing. The factors of the 52 items were
involving 35 parents. The test-retest analysis found that the determined through EFA, using Principal Axis Factoring and
ICC for all items was > 0.5, except for item D1, which had an Promax rotation. The Kaiser-Meyer-Olkin (KMO) value
ICC of 0.44, reflecting poor stability over time. This item was obtained was 0.882, and Bartlett’s test of sphericity was
significant, indicating that sampling adequacy was fulfilled

89
 ASM Science Journal, Volume 13, Special Issue 5, 2020 for APRU2018

and that the items were suitable to be factorised. Using Table 5. Factor Analysis and Internal Consistency of
Respondents’ Sociodemographic Characteristics
parallel analysis (Timmerman and Lorenzo-Seva, 2011), nine
factors were recommended to be extracted. The items were
rotated, and factor loadings were examined for poor loading
or cross-loading. Item G1 had poor loading and was deleted.
Then, the remaining 51 items were rotated again. The rotation
extracted 9 factors, accounting for 73.8% of the shared
variance. Examination of the pattern matrix revealed no
items that had poor factor loading or cross-loading issues.
Items B1‒B8 were loaded under a domain labelled ‘perceived
susceptibility’. Items C1‒C8 were loaded together and
labelled ‘perceived severity’. Items D2‒D7 were loaded under
a domain labelled ‘perceived self-efficacy’. Items E1‒E7 were
loaded together under a domain labelled ‘perceived response-
efficacy’.
Items F1, F4 and F5 were loaded together and labelled
‘perceived psychology cost’. Items F2, F3, F6 and F7 were
loaded under a domain labelled ‘perceived tangible cost’.
Items G2‒G5 were loaded together under a domain labelled
‘perceived maladaptive reward’. Items H1‒H4 were loaded
together under a domain labelled ‘discursive digital security
practice’, and items H5‒H11 were loaded together and
labelled ‘control digital security practice’ (Table 6).

Table 6. Factor Loading for Items According to Rotated Pattern Matrix

Component Item no Item Factor loading

Perceived susceptibility Be bullied (harassed, threatened and/or intimidated) online.

B1 0.74
(When your child is using the
internet, in your opinion, how Be spending more time online than he/she should.
likely or unlikely is it that B2 0.60
he/she will) Be exposed to adult content (e.g. pornography, violence, gambling).
B3 0.83
Have his/her personal information obtained without his/her knowledge or consent.
B4 0.84
Be approached online by a person he/she does not know.
B5 0.83
Exchange sexual messages and/or images with other people online.
B6 0.90
Be exposed to online content promoting self-harm (e.g. websites that encourage suicide,
B7 0.86
eating disorders, drug use).
Be exposed to online content that promotes hate, extreme views or terrorism.
B8 0.81
Perceived severity (In your A child is being bullied (harassed, threatened and/or intimidated) online.
C1 0.86
opinion, how serious are
these issues to you?) A child is spending more time online than he/she should.
C2 0.68
A child is exposed to adult content (including pornography, violence, gambling).
C3 0.89
C4 A child’s personal information is obtained without his/her knowledge or consent. 0.92

90
 ASM Science Journal, Volume 13, Special Issue 5, 2020 for APRU2018

A child is approached online by a person he/she does not know.

C5 0.91
A child exchanges sexual messages and/or images with other people online.
C6 0.95
A child is exposed to online content that promotes self-harm (e.g. websites that encourage
C7 suicide, promote eating disorders, drug use). 0.92

A child is exposed to online content that promotes hate, extreme views or terrorism.
C8 0.93
Perceived self-efficacy (How I am confident in my knowledge about keeping my child safe online.
D2 0.78
much do you agree or
disagree with the following I am comfortable with using the internet together with my child.
statement?) D3 0.56
I am confident in imposing rules on internet use on my child.
D4 0.91
I am confident in using filtering and monitoring software (parental control applications).
D5 0.67
I am comfortable with restricting my child to using the internet only when I am around.
D6 0.68
I am confident in checking my child’s online activities after my child has been online.
D7 0.54
Perceived response efficacy Discussing online safety with my child will keep him/her safe online.
E1 0.87
(How much do you agree or
disagree with the following Having the appropriate knowledge will keep my child safe online.
statement?) E2 0.91
Using the internet together with my child will keep him/her safe online.
E3 0.75
Imposing internet rules on my child will keep him/her safe online.
E4 0.94
Using filtering and monitoring software (parental control applications) will keep him/her
E5 safe online. 0.80

Restricting my child to using the internet only when I am around will keep him/her safe
E6 online. 0.70

Checking my child’s online activities after he/she has been online will keep him/her safe
E7 online. 0.70

Perceived psychological cost F1 Discussing online safety with my child is troublesome for me.
(How much do you agree or 0.65
disagree with the following F4 Ensuring that my child follows internet rules is troublesome for me. 0.85
statement?)
F5 Ensuring that filtering and monitoring software (parental control applications) are working 0.81
can be troublesome for me.

Perceived tangible cost (How F2 It takes a lot of effort to acquire appropriate knowledge about online safety.
0.74
much do you agree or
disagree with the following F3 It takes a lot of effort to use the internet together with my child.
statement?)
*effort refers to attempts to ensure using the internet together with your child (e.g. arranging 0.89
daily routines or setting up a calendar to schedule time to use the internet together).

F6 Restricting my child to using the internet only when I am around requires a lot of effort.

*effort refers to attempts to ensure that your child uses the internet only when a parent is 0.76
around (e.g. rules allowing use of devices only in common areas of the house).

F7 Checking my child’s online activities after he/she has been online requires a lot of effort.
0.77

Perceived maladaptive Allowing my child to use the internet on his/her own will enable me to focus on my own
reward (How much do you G2 interests. 0.65
agree or disagree with the
following statement?) Not imposing internet rules on my child will make him/her happy.
G3 0.83
By not putting up filtering and monitoring software (parental control applications), my
G4 child can use the internet freely. 0.76

91
 ASM Science Journal, Volume 13, Special Issue 5, 2020 for APRU2018

By not checking my child’s online activities after he/she has been online, I am respecting
G5 his/her rights. 0.52

Discursive security practice Discuss online safety with your child.

H1 0.65
(How often do you)
Have conversations with your child on how to handle unknown people online.
H2 0.86
Discuss with your child how to protect personal information online.
H3 0.89
Have conversations on what to do if he/she is bullied or harassed online.
H4 0.81
Control security practice Use the internet together with your child.
H5 0.64
(How often do you)
Tell your child when/how long to use the internet.
H6 0.76
Tell your child which websites/social networks he/she can visit.
H7 0.73
Tell your child what he/she can and cannot do online.
H8 0.61
Ensure that filtering and monitoring software (parental control applications) are present.
H9 0.53
Restrict your child to using the internet only when you are present.
H10 0.84
Check the websites that your child has visited.
H11 0.77

The internal consistency of the factors produced was IV. DISCUSSION

examined. All the factors exhibited good internal consistency
(Table 7) with their respective items, as recommended in the In this study, we developed the P-Dis questionnaire to assess
literature (Mokkink et al., 2010). The inter-item correlation parents’ digital-security practices aimed at keeping their
of each domain ranged from 0.3‒0.9. The corrected item- children safe online. The final questionnaire consisted of 51
total correlation (CITC) was > 0.3, and Cronbach’s alpha was > items covering 9 components. The internal consistency
0.7. reliability of each component was good, with the Cronbach’s
alpha coefficient ranging from 0.79‒0.97. The components
Table 7. Internal Consistency Summary of the Factors were derived based on factor analysis and represented the
themes collected from the PMT framework, discussions with
experts, review of literature and an online survey of parents.
As the questionnaire developed was heavily based on the
PMT framework, the factors extracted were interpreted and
compared with the underlying PMT domains whenever
possible. The items under the ‘perceived susceptibility’,
‘perceived severity’, ‘perceived self-efficacy’, ‘perceived
response efficacy’ and ‘perceived maladaptive reward’ factors
were consistent with the domains they were intended to
represent and were in line with PMT domains. ‘Perceived
psychological cost’ and ‘perceived tangible cost’ are new
factors discovered in this EFA. ‘Perceived psychological cost’
was labelled as such because the items that loaded into this
factor had a common keyword, ‘troublesome’, which reflects
the psychological state while performing the actions
described in those items. ‘Perceived tangible cost’ was
labelled as such because the items that loaded into this factor
had a common keyword, ‘effort’, which reflects measurable

92
 ASM Science Journal, Volume 13, Special Issue 5, 2020 for APRU2018

costs, including time and physical action. Both ‘perceived trust in their children and their levels of understanding and
tangible cost’ and ‘perceived psychological cost’ can be information on internet-based devices (Meehan S., 2016).
treated as formative components of ‘perceived response cost’, Parental control mediation is similar to ‘control digital
based on theoretical argument and content validation. From security’ in the present study, and ‘parental experience’ is
the theoretical aspect, PMT shares certain similarities with reflected in the present study’s ‘discursive digital security’
another cognitive-based model, the Health Belief Model practice. Hence, the formation of these two factors is
(Prentice-Dunn and Rogers, 1986). Rogers et al. explicitly justifiable based on the literature and the validation of the
mentioned that the component of response cost was content by experts. Overall, the factors that emerged from the
equivalent to the model’s perceived barriers (Prentice-Dunn feedback received from both experts and parents were
and Rogers, 1986). Perceived barriers in the model were deemed sufficient to understand parental digital-security
defined as ‘Beliefs about the tangible and psychological costs practice, and the items were relevant and well received by
of the advised action’ (Glanz et al., 2008). Hence, this parents.
definition supports the argument that ‘perceived This study has some strengths that merit highlighting. First,
psychological cost’ and ‘perceived tangible cost’ are in line this is the first validated questionnaire that measures
with the underlying theoretical definition and justify being parental digital-security practices from the well-established,
labelled as such. ‘Discursive digital security practice’ was cognitive-based PMT in the Malaysian context. The
labelled as such because the items that loaded into this factor development of the questionnaire was based on best practices
reflected active and discussion-based actions. ‘Control digital and underwent comprehensive measures from item
security practice’ was named as such because the items that development and scale development to testing the
loaded into this factor reflected a common theme of exertion psychometric properties. The questionnaire produced is dual
of power and authority by parents in performing the actions. language, increasing its potential for use among the
Based on content validity, the experts agreed that the items Malaysian population.
forming these two factors reflected digital-security practices Although the P-Dis questionnaire fulfilled its psychometric
in general. The literature also supports the grouping of these properties, there are a few limitations to the questionnaire
two types of digital-security practices. Wisniewski et al. have that need to be highlighted. The majority of the parents who
described two types of parental mediation practices regarding participated were from the central region of Malaysia, and
social media use by their children: direct and active hence the study’s findings might not represent the parent
mediation (Wisniewski et al., 2015). According to the authors, populations in other regions of Malaysia. Similarly, although
direct interventions include actions taken by parents to there were respondents from lower social statuses, the
directly intervene in their children’s social media use, majority of the respondents were had higher levels of
including applying rules and restrictions (Wisniewski et al., education and employment. Therefore, the questionnaire’s
2015). Active mediation applies when parents take actions validity needs to be replicated in other regions of Malaysia
that include talking to their children but not attempting to and with certain population demographic characteristics,
directly control their social media use (Wisniewski et al., including lower social statuses. Nevertheless, this study
2015). These two types of digital-security practices described provides a significant initial step toward introducing a
by Wisniewski et al. (2015) are similar to the ‘discursive validated instrument regarding parental digital-security
digital security’ and ‘control digital security’ practices in the practices in the Malaysian setting.
present study. A qualitative study by Meehan (2016) also
highlighted two types of parental mediation strategies for V. CONCLUSION
managing children’s use of internet-connected devices:
control and parental experience. Parental control mediation The 51-item P-Dis questionnaire underwent robust steps in
includes ‘covert and overt strategies and tactics’, whereas its development and was proved to fulfil its psychometric
parental experience is associated with the level of parents’

93
 ASM Science Journal, Volume 13, Special Issue 5, 2020 for APRU2018

properties. As such, the questionnaire can be used in future for parents and that efficiently improve their understanding
research, particularly that involving the Malaysian and knowledge of cyber-parenting in general.
population. For example, the questionnaire can be used to
understand the role of parental digital-security practices VI. ACKNOWLEDGEMENTS
regarding certain online behaviours concerning public health,
including cyberbullying and internet addiction. In practical The authors would like to thankfully acknowledge the support
terms, the questionnaire’s ability to identify the factors and shown by collaborative actions by Cyber Security Malaysia,
practices used by parents to keep their children safe online the University of Malaya and Ministry of Health Malaysia,
may help stakeholders provide interventions that are suitable which made this study a success.

VII. REFERENCES

Beaton, D. E., Bombardier, C., Guillemin, F. & Ferraz, M. B. Cyberpsychology, Behavior, and Social Networking, vol. 20,
2000, ‘Guidelines for the Process of Cross-Cultural pp. 362-368.
Adaptation of Self-Report Measures’, Spine (Phila Pa 1976), Hwang, Y. & Jeong, S.-H. 2015, ‘Predictors of Parental
vol. 25, pp.3186-91. Mediation Regarding Children's Smartphone Use’,
Cortina, J. M. 1993, ‘What is Coefficient Alpha? An Cyberpsychology, Behavior, and Social Networking, vol. 18,
Examination of Theory and Applications’, Journal of pp. 737-743.
Applied Psychology, vol. 78, pp. 98. Koo, T. K. & Li, M. Y. 2016, ‘A Guideline of Selecting and
Cristobal, E., Flavián, C. & Guinaliu, M. 2007, ‘Perceived e- Reporting Intraclass Correlation Coefficients for Reliability
Service Quality (Pesq) Measurement Validation and Effects Research’, Journal of Chiropractic Medicine, vol. 15, pp.
on Consumer Satisfaction and Web Site Loyalty’, Managing 155-163.
Service Quality: An International Journal, vol.17, pp. 317- Lorenz, B. 2017, ‘A Digital Safety Model for Understanding
340. Teenage Internet User’s Concerns’, PhD Thesis, Tallinn
Cyber Security Malaysia, 2014, Cybersafe in Schools- A University.
National Survey Report, Malaysian Communications and Lynn, M. R. 1986, ‘Determination and Quantification of
Multimedia Commission. Content Validity’, Nursing Research.
Di Lorio, C. K. 2005, Measurement in Health Behaviour, San Maddux, J. E. & Rogers, R. W. 1983, ‘Protection Motivation
Francisco, John Wiley & Sons. Theory and Self-Efficacy: A Revised Theory of Fear Appeals
Glanz, K., Rimer, B. K. & Viswanath, K. 2008, Health and Attitude Change’, Journal of Experimental Social
Behavior and Health Education: Theory, Research, and Psychology, vol. 19, pp. 469-479.
Practice, John Wiley & Sons. Meehan, S 2016, ‘A Model of Parental Mediation of their
Guillemin, F., Bombardier, C. & Beaton, D, 1993, ‘Cross- Children’s Use of Internet Connected Devices’, Psychol Clin
Cultural Adaptation of Health-Related Quality of Life Psychiatry, vol 5.
Measures: Literature Review and Proposed Guidelines’, Mokkink, L. B., Terwee, C. B., Patrick, D. L., Alonso, J.,
Journal of Clinical Epidemiology, vol. 46, pp. 1417-1432. Stratford, P. W., Knol, D. L. & De Vet, H. C. W. 2010, ‘The
Hair, J., Black, W. & Babin, B. 2010, Anderson. Re, 2010, Cosmin Checklist for Assessing the Methodological Quality
Multivariate Data Analysis, New Jersey, Pearson Prentice of Studies on Measurement Properties of Health Status
Hall. Measurement Instruments: An International Delphi Study’,
Hinkin, T. R. 1995, ‘A Review of Scale Development Practices Quality Of Life Research, vol. 19, no.14.
in The Study of Organizations’, Journal of Management, vol.
21, pp. 967-988. Nathanson, A. I. 2001, ‘Parent and Child Perspectives on the
Hwang, Y., Choi, I., Yum, J.-Y. & Jeong, S.-H. 2017, ‘Parental Presence and Meaning of Parental Television Mediation’,
Mediation Regarding Children's Smartphone Use: Role of Journal of Broadcasting & Electronic Media, vol. 45, pp.
Protection Motivation and Parenting Style’, 201–220.

94
 ASM Science Journal, Volume 13, Special Issue 5, 2020 for APRU2018

Nikken, P. & Jansz, J. 2014, ‘Developing Scales to Measure

Parental Mediation of Young Children's Internet Use’,
Learning Media and Technology, vol. 39, pp. 250-266.
Polit, D. F. & Beck, C. T. 2006, ‘The Content Validity Index:
Are You Sure You Know What's Being Reported? Critique
and Recommendations.’, Res Nurs Health, vol. 29, pp. 489-
97.
Prentice-Dunn, S. & Rogers, R. W. 1986, ‘Protection
Motivation Theory and Preventive Health: Beyond the
Health Belief Model’, Health Education Research, vol. 1, pp.
153-161.
Sonck, N., Nikken, P. & De Haan, J. 2013, ‘Determinants of
Internet Mediation: A Comparison of The Reports by Dutch
Parents and Children’, Journal of Children and Media, vol.
7, pp. 96-113.
Streiner, D. & Norman, G. 2008, Health Measurement Scales:
A Practical Guide to their Development and Use, Oxford
University Press.
Tafforeau, J., Cobo, M. L., Tolonen, H., Scheidt-Nave, C. &
Tinto, A. 2005, Guidelines for the Development and Criteria
for the Adoption of Health Survey Instruments, In:
Statistics, O. F. (Ed.). Luxembourg: European Commission.
Timmerman, M. E. & Lorenzo-Seva, U. 2011, ‘Dimensionality
Assessment of Ordered Polytomous Items with Parallel
Analysis’, Psychol Methods, vol. 16, pp. 209-20.
UNICEF Malaysia, 2014. Exploring the Digital Landscape in
Malaysia Access and Use of Digital Technologies by
Children and Adolescents, UNICEF Malaysia.
Vitoratou S, Ntzoufras I, Smyrnis N & Stefanis, N. 2009,
‘Factorial Composition of The Aggression Questionnaire: A
Multi-Sample Study in Greek Adults’, Psychiatry Res., vol.
168.
Willingham, D. T. 2007, Cognition: The Thinking Animal,
Pearson/Prentice Hall Englewood Cliffs, Nj.
Wisniewski, P., Jia, H., Xu, H., Rosson, M. B. & Carroll, J. M.
2015, ‘Preventative Vs. Reactive: How Parental Mediation
Influences Teens' Social Media Privacy Behaviors’, in
Proceedings of the 18th ACM Conference on Computer
Supported Cooperative Work & Social Computing, 2015.
Worthington, R. L. & Whittaker, T. A. 2006, ‘Scale
Development Research: A Content Analysis and
Recommendations for Best Practices’, The Counselling
Psychologist, vol. 34, pp. 806-838.

95