Mobile app impulsive buying: A situational factors dataset analysis

This dataset examines the empirical model of situational factors for impulsive buying on mobile apps. The model presents factors driving impulse buying based on situational factors consisting of the physical environment, social environment, time perspective, and hedonic and utilitarian exploration. The data collection technique in this study used a questionnaire distributed online according to predetermined criteria, namely mobile device users who accessed online market board applications and made regular purchases in the last six months. The data were tested statistically using Structural Equation Modeling to ensure the model's fit.


Subject
Business, Management, and decision sciences.

Specific subject area Marketing Data format
Raw and analyzed.Type of data Excel Spreadsheet and Tables.

Data collection
Data were obtained with a survey through a questionnaire regarding the effect of physical environment, social environment, time perspective, and hedonic-utilitarian browsing on mobile apps impulse buying.

Data source location
The data was obtained on the Indonesian online market using an online survey.

Value of the Data
• These datasets provide valuable insight into the critical factors for predicting the driving factors of impulse buying on mobile apps, which consists of the physical environment, social environment, time perspective, and hedonic and utilitarian browsing.The data meets the model eligibility rules to describe the suitability of empirical conditions and the research model.The data provides an overview of empirical conditions for understanding impulse buying on mobile apps.This data survey can be evaluated by replicating research in other cultures or regions.• This data also provides essential information on digital marketers considering appropriate online marketing communication programs.Understanding the driving factors of impulse buying on mobile apps can provide relevant information for digital marketing practitioners.• This data is valuable for researchers interested in digital marketing on mobile apps.Model test data provide evidence of a model that can be developed or replicated for future research focusing on online impulsive buying.

Data Description
These datasets provide the critical factor for predicting the driving factors of impulse buying on mobile apps, which consists of physical environment, social environment, time perspective, and hedonic and utilitarian browsing.The data were collected using a questionnaire, available as a supplementary file adapted from [ 1 , 2 , 3 ] and stored in the Mendeley repository.The data contains the results of respondents' statements with a scale of 1-5 from strongly disagree to agree strongly.Each indicator represents each variable with a symbol: PE (physical environment), which consists of 4 items, namely PE1, PE2, PE3, and PE4, SE (social environment) which consists of 3 items namely SE1, SE2, and SE3, TP (time perspective) which consists of 3 items namely TP1, TP2, and TP3, HB (hedonic browsing) which consists of 3 items namely HB1, HB2, and HB3, UB (utilitarian browsing) which consists of 5 items namely UB1, UB2, UB3, UB4 and UB5, and (IB) impulse buying which consists of 3 items namely IB1, IB2, and IB3.A summary of each variable measurement data is presented in Table 1 .Structural equation modelling ensures that the data meets the rules of fit or is appropriate based on the goodness of fit indicator.The data regarding the sample demographics are presented in Table 2 .The data were analyzed through the data quality test stage.The manifest variable data fulfils the normality test by having a critical ratio below ± 2.58.The data were also tested by using the multivariate outlier data using limit 4 [4] .Based on the multivariate outliers test, it is known that there are no outliers in all variables indicated by the Mahalanobis distance value, which is available as a supplementary file.Table 3 presents the factor load value measured from the latent variable through each observed variable.In making decisions regarding the suitability of latent variables with the observed variables, the minimum factor loading value criteria are set at 0.5 [5] .Overall, the factor loading value of all observed variables from the latent variables is valid and meets the methodological fit measurement model criteria.Reliability is needed to measure the internal consistency of indicators of a construct.Tests of the instrument are carried out using composite reliability and variance extracted.Table 3 shows that the composite reliability coefficient for each variable is above the acceptance value of the reliability limit, namely the minimum value of 0.7 [4] .The value of variant extraction is at the acceptance limit level, a minimum value of 0.5.The test data for the variable measurement model are presented in Table 4 .Based on the measurement model testing, the Chi-square, CMIN/df, RMSEA, GFI, TLI, and CFI values meet the required values.

Experimental Design, Materials and Methods
The data were collected using a questionnaire with a Likert scale.The questionnaire was adapted from previous research [1][2][3] by conducting a pilot test with as many as 60 respondents or consumers who are used to making online purchases to make the instrument feasible.Structural Equation Modeling is used to measure the acceptance of the empirical model.Structural equation modelling has the advantage of integrating factor analysis with path analysis [4] .The samples collected were 628 respondents from several online platforms in Indonesia.The sampling technique uses non-probability with purposive sampling method, with the criteria of Indonesian consumers aged 18-40 years who were considered to have mobile devices.Data was collected for six months, from April to October 2022, and transformed into an Excel spreadsheet.Data was collected by paying attention to the ethical principles of data collection.Respondents have explained the purpose of data collection.Respondent data is also kept confidential for the privacy of data sources.Data were processed statistically, assisted by the AMOS.

Table 1
Construct and measurement.

Table 3
Validity and reliability testing.

Table 4
Variable's Goodness of fit.