Data modeling positive security behavior implementation among smart device users in Indonesia: A partial least squares structural equation modeling approach (PLS-SEM)

The article presents raw inferential statistical data related to understanding the positive security behaviors of smart device users in Indonesia, which was used to determine whether the studied variables were direct or mediating factors. The factors explored include government efforts, technology provider support, privacy concerns, trust, perceived behavioral control, attitudes, and subjective norms. The theory of planned behavior was adopted to develop the proposed model for implementing positive security behaviors. Structured questionnaires were distributed via an online survey to consumers currently using a smartphone or using a smartphone and some other smart device. Furthermore, the respondents were from 19 provinces in Indonesia. The quantitative research method was used to analyze the data. Reliability and validity were confirmed. Structural equation modeling (SEM) using the Smart PLS software version 3 was used to present data. SEM path analysis identified estimates of the relationships of the primary constructs in the data. The outcomes obtained from this dataset demonstrate a direct influence between government efforts, privacy, and perceived behavioral control and performing positive security behaviors. Other variables had positive and significant influences on implementing positive security behaviors, indicating their roles as mediation variables. This data is useful for reference and consideration in the improvement of smart device users’ security behaviors. This data can also provide valuable insights to countries with characteristics that are similar to those of Indonesia.


a b s t r a c t
The article presents raw inferential statistical data related to understanding the positive security behaviors of smart device users in Indonesia, which was used to determine whether the studied variables were direct or mediating factors. The factors explored include government efforts, technology provider support, privacy concerns, trust, perceived behavioral control, attitudes, and subjective norms. The theory of planned behavior was adopted to develop the proposed model for implementing positive security behaviors. Structured questionnaires were distributed via an online survey to consumers currently using a smartphone or using a smartphone and some other smart device. Furthermore, the respondents were from 19 provinces in Indonesia. The quantitative research method was used to analyze the data. Reliability and validity were confirmed. Structural equation modeling (SEM) using the Smart PLS software version 3 was used to present data. SEM path analysis identified estimates of the relationships of the primary constructs in the data. The outcomes obtained from this dataset demonstrate a direct influ-ence between government effort s, privacy, and perceived behavioral control and performing positive security behaviors. Other variables had positive and significant influences on implementing positive security behaviors, indicating their roles as mediation variables. This data is useful for reference and consideration in the improvement of smart device users' security behaviors. This data can also provide valuable insights to countries with characteristics that are similar to those of Indonesia.
© 2020 The Author(s). Published by Elsevier Inc. This is an open access article under the CC BY license.
( http://creativecommons.org/licenses/by/4.0/ ) Specifications Table   Subject Computer science (general) Specific subject area Information system Type of data

Value of the Data
• The data is useful for all stakeholders, such as technology providers, academicians, especially the government of Indonesia, in terms of improving security awareness effort s among smart device users.
• The data presents how government efforts, technology provider support, trust, privacy concerns, attitudes, subjective norms, and perceived behavioral control impact smart device users' positive security behaviors. This information is useful because it can serve as a reference and be considered in the development of measures to improve smart device users' security behaviors. • This data can be used to develop a measurement tool to determine the positive security behaviors related to the use of smart devices in another context. • This data can provide useful insights for countries with characteristics that are like those of Indonesia.

Data Description
The facts and statistics presented in this paper were collected via primary data collection through an online survey, which can be accessed at the following link: s.id/privasiperangkatpintar (in Bahasa). The questionnaire in English is provided as a Supplementary File. The researchers developed the survey instrument using research constructs based on previous studies, as shown in Table 1 .
The wording of the questionnaire was initially developed in English and then translated into the local language (Bahasa). The survey was divided into two parts. Part A addressed demographic information, including respondents' age, gender, educational qualifications, and smart device ownership. Part B included questions covering the different constructs in the proposed research model using a five-point Likert scale ranging from (1) "strongly disagree" to (5) "strongly agree." The online communication channel, namely WhatsApp, was used to distribute the questionnaire. After eliminating invalid responses, that is, 18 respondents filled incomplete questionnaires; data from 314 respondents were analyzed. The demographic characteristics of the respondents are shown in Table 2 .
The graph in Fig. 1 shows the kinds of smart devices owned by the respondents. Among the 106 respondents with a smart device besides a smartphone, 78 respondents (around 73.6%) owned a smart TV. Furthermore, the chart in Fig. 2 reveals that about 7.7% of the 106 respondents had more than one type of smart device.

Experimental Design, Materials, and Methods
The presented data were collected based on quantitative research methods. A survey method was chosen as the preferred technique because it provides many benefits, including allowing the collection of standardized data, which enabled researchers to meet the aim of the research [ 9 , 10 ], namely, understanding the factors that influence smart device users' positive security behaviors.
Current smart device users and smartphone users, who were assumed to be potential adopters of other smart devices in some regions in Indonesia, were selected as respondents.
The researchers proposed a model to test the data. The model consists of constructs: government efforts, technology provider support, trust, and privacy, as well as attitudes, subjective norms, and perceived behavioral control, could directly influence positive security behavior or serve as mediation variables to influence positive security behavior. The quality of the measurement model was determined based on its validity and reliability by considering the following values: Cronbach's alpha ( > 0.60), composite reliability ( > 0.70), average variance extracted (AVE)( > 0.50), and loading factor (0.70) [11] . The measurement accuracy data can be seen in Table 3 . -Positive security behavior 1. Reading the privacy policy statement carefully before using the device is important. [1] 2. I know where to report an incident related to smart devices' security. 3. I know of privacy issues related to the use of smart devices that I have. 4. I know how to control the personal information given to smart devices. 5. I can control the protection of my personal information on all smart devices that I have.   The proposed research model was used to empirically analyze the data using the partial least squares structural equation modeling (PLS-SEM) technique, and SmartPLS version 3 software was used to code the data and run the statistical analysis. PLS-SEM is known to be reliable in sample distribution and small sample size. The structural model can be seen in Fig. 3 .
The structural model was examined by testing the hypothesized relationships. Moreover, the bootstrapping method was used on 5,0 0 0 subsamples to assess the significance and path coefficients, as suggested by Hair et al. [12] . The output model analysis data is displayed in Table 4 .    Fig. 3. Measurement and structural model analysis.

Ethical considerations
The researchers ensured that respondents were well informed about the background and the aim of this research. Respondents were also assured of the confidentiality of the data they submitted in the survey.

Academic, practical, and policy implications of this data article
The data presented in this article offers implications for the academic field. Some variables directly influenced users' performance of positive security behaviors, while other variables had positive and significant values, indicating their roles as mediation variables. For example, among the constructs of the theory of planned behavior, the data indicates that only perceived behavioral control directly influenced positive security behavior, given the strong relationship between them ( β = 0.537). Meanwhile, subjective norms influenced attitudes ( β = 0.215) and perceived behavioral control ( β = 0.127) in a positive and significant way, and attitudes influenced perceived behavioral control with a path coefficient of β = 0.166. Therefore, among academics in the security awareness field, this finding can enhance understanding of how mediation variables can lead users actually to perform positive security behaviors.
The data also indicates that government effort s directly influenced positive security behavior in a positive and significant way, as indicated by a path coefficient of β= 0.139. Based on Fig. 3 , R 2 demonstrates that the research model explains 47.9% of the variance in performing positive security behavior. Furthermore, the data indicates that government effort s influenced perceived behavioral control in a positive and significant way with a path coefficient of β = 0.151. The present findings also note there are more people use more than one smart device in addition to their smartphone. Regarding practical implications, the data presented in this article can help policymakers who are developing security policies enhance users' positive security behaviors. Overall, insights from this dataset can be used to create new strategies and guide the revision of existing policies.

Acknowledgments
This data article was supported by Doctoral Grant Universitas Indonesia 2020. The authors also express sincere appreciation to the Centre for Research and Development of Postal and Information Technology Resources, Equipment and Services (PITRES) 2020, Research and Human Resources Development Agency, Ministry of Communication and Informatics of the Republic of Indonesia for partially supported for the publication of this work. Moreover, we would like to gratefully acknowledge the insightful review and suggestion from the reviewer on the earlier version of this article.

Conflict of Interest
I, Kautsarina, and my colleagues write to declare that there is no conflict of interest traceable to our data paper "Data modeling positive security behavior implementation among smart device users in Indonesia: A partial least squares structural equation modeling approach (PLS-SEM)".

Supplementary materials
Supplementary material associated with this article can be found, in the online version, at doi: 10.1016/j.dib.2020.105588 .