Links to Data and Statistical Analysis Repository

Data and Analysis Underlying Study of Factors Affecting User's Behavioral Intention and Use of a Mobile-Phone-Delivered Cognitive Behavioral Therapy for Insomnia

Siska Fitrianie, Corine Horsch, Robbert Jan Beun, Fiemke Griffioen-Both, Willem-Paul Brinkman

Corresponding author: Siska Fitrianie, s.fitrianie@gmail.com
https://doi.org/10.4121/16825843

This document provides links to files underlying the analysis presented in the article:

Siska Fitrianie, Corine Horsch, Robbert Jan Beun, Fiemke Griffioen-Both and Willem-Paul Brinkman. Factors Affecting User’s Behavioral Intention and Use of a Mobile-Phone-Delivered Cognitive Behavioral Therapy for Insomnia: A Small-Scale UTAUT Analysis. Journal of Medical Systems 45, 110 (2021). https://doi.org/10.1007/s10916-021-01785-w.

The analysis was carried out on data gathered from a field trial involving people (n = 89) with relatively mild insomnia using a Cognitive Behavioral Therapy of-Insomnia (CBT-I) app. It studied an extended version of the Unified Theory of Acceptance and Use of Technology (UTAUT2) model to examine factors influencing users' behavioral intention to use a CBT-I app and their app-use behavior. The Partial Least Squares-Structural Equation Modeling method was applied.

Below is the table of content (TOC) of the article. Within this TOC, links to codes, data and the output(s) of the executed codes are presented. These codes and data are necessary for calculations or analyses that are presented in the paper. The outputs of the calculations or analyses are presented in tables, figures or within the text of the corresponding sections.

Requirements

1. SmartPLS to run smartpls codes
2. Ms Excel 2003 or higher to open the xlxs files of the smartpls results
3. R version 3.3.3 or higher to run R codes

Table of Content:

• Abstract
  

• Introduction
  

• Research Model
  

• Method
  
  • Participant
    
    • Table 3 Demographic characteristics of participants participated in the Sleep App Acceptance Questionnaire
      • codes: participants.R
      • Required data files: 89_beforeimputation.sav
      • output: participants.txt
  • System Description
    
  • Measurement
    
    • Questionnaire Measures
      
    • Behavioral Measures
      
      • Source: behavioral_calc.R
      • Required data file: appData.sav
      • output: (append to) appData.sav
      • Note: data (n = 89) before the imputation process was created as the subset of combination of appData.sav and acceptance_rawData.sav.
      • Variable description: variableNames.pdf.
  • Procedure
    
  • Data Analysis
    
    • Minimum Sample Size
      
    • Data Characteristics
      
      • Variability of Constructs
        • Source: variability_aftercomputation.R
        • Required data files (n = 89) before imputation: 89_imputed.sav
        • output: variability_aftercomputation.txt
      • Missing Values
        • Source: missing_data.R
        • Required data files (n = 89) before imputation: 89_beforeimputation.sav
        • output: description_missingdata.txt
      • The imputation of Missing Values
        • Source: mice_imputation.R
        • Required data files: 89_beforeimputation.sav
        • output: CSV format 89_imputed.csv or SAV format 89_imputed.sav
    • Latent Variable Assessment
      • Note: this assessment of reflective variables was conducted based on the multipath-test and the results was used as inclusion/exclusion criteria of indicators
      • Source: SmartPLS 3.0 codes and data (Intention-Usage)
      • SAV data: 89_imputed.sav
      • Variable names: variable_names.txt
      • Output in EXCEL files: Intention-Usage.xlsx (Bootstrap) and Intention-Usage.xlsx (PLS Algorithm)
      • Individual outputs of Performance Expectancy (PE), Effort Expectancy (EE), Social Influence (SI), Self-Efficacy (SE), Trust (TR), Hedonic Motivation (HM), Anxiety (AN), Facilitating Condition (FC), Behavioral Intention (BI), and Use Behavior (UB):
        • Outer Loadings
        • Cronbach's α
        • AVE
        • Discriminant Validity (Cross Loadings)
  • Results
    
    • Single-Path Tests
      • Table 6 The bootstrapping results of the single-path analyses (n = 89)
        • Behavioral Intention (BI) as dependent variable
          • Sources: SmartPLS 3.0 codes and data (BI as DV)
          • SAV data: 89_imputed.sav
          • Variable names: variable_names.txt
          • Output in EXCEL files: PLS-SEM analysis (BI as DV) and Cross-Validation
          • Individual outputs:
            • H₁: Performance Expectancy (PE)
              • R-square
              • f-square
              • Path coefficient (with the 95% confidence intervals based on on Bias-Corrected and Accelerate (BCa))
              • Q-square
            • H₂: Effort Expectancy (EE)
              • R-square
              • f-square
              • Path coefficient (with the 95% confidence intervals based on on Bias-Corrected and Accelerate (BCa))
              • Q-square
            • H₃: Social Influence (SI)
              • R-square
              • f-square
              • Path coefficient (with the 95% confidence intervals based on on Bias-Corrected and Accelerate (BCa))
              • Q-square
            • H₄: Self-Efficacy (SE)
              • R-square
              • f-square
              • Path coefficient (with the 95% confidence intervals based on on Bias-Corrected and Accelerate (BCa))
              • Q-square
            • H₅: Trust (TR)
              • R-square
              • f-square
              • Path coefficient (with the 95% confidence intervals based on on Bias-Corrected and Accelerate (BCa))
              • Q-square
            • H₆: Hedonic Motivation (HM)
              • R-square
              • f-square
              • Path coefficient (with the 95% confidence intervals based on on Bias-Corrected and Accelerate (BCa))
              • Q-square
            • H₇: Anxiety (AN)
              • R-square
              • f-square
              • Path coefficient (with the 95% confidence intervals based on on Bias-Corrected and Accelerate (BCa))
              • Q-square
            • H₈: Facilitating Conditions (FC)
              • R-square
              • f-square
              • Path coefficient (with the 95% confidence intervals based on on Bias-Corrected and Accelerate (BCa))
              • Q-square
        • Use Behavior (UB) as dependent variable
          • Source: SmartPLS 3.0 codes and data (UB as DV)
          • SAV data: 89_imputed.sav
          • Variable names: variable_names.txt
          • Output in EXCEL files: PLS-SEM Analysis (UB as DV) and Cross-Validation (UB as DV)
          • Individual outputs:
            • H₉: Behavioral Intention (BI)
              • R-square
              • f-square
              • Path coefficient (with the 95% confidence intervals based on on Bias-Corrected and Accelerate (BCa))
              • Q-square
            • H₁₀: Facilitating Condition (FC)
              • R-square
              • f-square
              • Path coefficient (with the 95% confidence intervals based on on Bias-Corrected and Accelerate (BCa))
              • Q-square
    • Multiple-Path Test
      • Figure 3 The bootstrapping results of the multiple-path analysis (n = 89).
        • Source: SmartPLS 3.0 codes and data (Intention-Usage)
        • SAV data: 89_imputed.sav
        • Variable names: variable_names.txt
        • Output in EXCEL files: Intention-Usage.xlsx and Intention-Usage_Q2.xlsx
        • Individual outputs of Behavioral Intention (BI) and Use Behavior (UB) as dependent variables:
          • R-square
          • f-square
          • Path coefficient (with the 95% confidence intervals based on on Bias-Corrected and Accelerate (BCa))
          • Outer Loadings
          • Q-square for Dependent Variables BI and UB
          • Q-square value of Dependent Variables by excluding significant Independent Variables:
            • Q-square of Behavioral Intention (BI) excluding Hedonic Motivation (HM)
            • Q-square of Use Behavior (UB) excluding Behavioral Intention (BI)
            • Q-square of Use Behavior (UB) excluding Facilitating Condition (FC)
            • q-square calculation:
              • R-script: small_q.R
              • SAV data: summary_Q2.sav
              • Results for HM-BI, BI-UB and FC-UB (n = 89): small_q2_89samples.txt
    • Mediation Analysis
      • Figure 4 Effects of the relationship between Hedonic Motivation and Use Behavior with mediation (n = 89)
        • Source: SmartPLS 3.0 codes and data (mediation_HM-BI-UB)
        • SAV data: 89_imputed.sav
        • Variable names: variable_names.txt
        • Output in EXCEL files: PLS-SEM Analysis (mediation HM-BI-UB)
        • Note: The Intention-Behavior Model is also referred "Intention-Usage" in the SmartPLS 3.0 codes and data
        • Individual outputs (with the 95% confidence intervals based on on Bias-Corrected and Accelerate (BCa)): Hedonic Motivation (HM) - Behavioral Intention (BI) - Use Behavior (UB)
          • βa = effect of HM-BI
          • βb = effect of BI-UB
          • βc = direct effect HM-UB without mediation
          • βaxb= indirect effect HM-UB (βa x βb)
          • βc' = direct effect HM-UB with mediation (βc - βaxb)
  • Discussion
    
  • Conclusions
    
  • Compliance with Ethical Standards
    
  Funding
  Etchical Approval
  Informed Consent
  Conflict of Interest
  
  • References
    
  • Appendix A Sleep App Acceptance Questionnaire
    • Sleep App Acceptance Questionnaire Items
  • Appendix B Data, Analysis, and Results of the Multi-Path Analysis
    
    • Description Analysis of Preprocessed Data
      • Source: description_beforeimputation.R
      • Required data files (n = 89) before imputation: 89_beforeimputation.sav
      • output: description_beforeimputation.txt
    • Summary of Variability and Assessment Results
      • Variability of Constructs
        • Source: variability_aftercomputation.R
        • Required data files (n = 89) before imputation: 89_imputed.sav
        • output: variability_aftercomputation.txt
      • Missing Values
        • Source: missing_data.R
        • Required data files (n = 89) before imputation: 89_beforeimputation.sav
        • output: description_missingdata.txt
      • The imputation of Missing Values
        • Source: mice_imputation.R
        • Required data files: 89_beforeimputation.sav
        • output: CSV format 89_imputed.csv or SAV format 89_imputed.sav
      • Assessment Result
        • See PLS Algorithm Results of the Multiple-Path Analysis (below)
    • PLS Algorithm Results of the Multiple-Path Analysis
      • Note: this assessment was conducted based on the multipath-test and the results was used as inclusion/exclusion criteria of indicators
      • Source: SmartPLS 3.0 codes and data (Intention-Usage)
      • SAV data: 89_imputed.sav
      • Variable names: variable_names.txt
      • Output in EXCEL files: Intention-Usage.xlsx (Bootstrap) and Intention-Usage.xlsx (PLS Algorithm)
      • Individual outputs of Performance Expectancy (PE), Effort Expectancy (EE), Social Influence (SI), Self-Efficacy (SE), Trust (TR), Hedonic Motivation (HM), Anxiety (AN), Facilitating Condition (FC), Behavioral Intention (BI), and Use Behavior (UB):
        • Outer Loadings
        • Cronbach's α
        • AVE
        • Discriminant Validity (Cross Loadings)
    • Bootsrapping Results of the Multiple-Path Analysis (Figure 3)
      • Source: SmartPLS 3.0 codes and data (Intention-Usage)
      • SAV data: 89_imputed.sav
      • Variable names: variable_names.txt
      • Output in EXCEL files: Intention-Usage.xlsx and Intention-Usage_Q2.xlsx
      • Individual outputs of Behavioral Intention (BI) and Use Behavior (UB) as dependent variables:
        • R-square
        • f-square
        • Path coefficient (with the 95% confidence intervals based on on Bias-Corrected and Accelerate (BCa))
        • Outer Loadings
        • Q-square for Dependent Variables BI and UB
        • Q-square value of Dependent Variables by excluding significant Independent Variables:
          • Q-square of Behavioral Intention (BI) excluding Hedonic Motivation (HM)
          • Q-square of Use Behavior (UB) excluding Behavioral Intention (BI)
          • Q-square of Use Behavior (UB) excluding Facilitating Condition (FC)
          • q-square calculation:
            • R-script: small_q.R
            • SAV data: summary_Q2.sav
            • Results for HM-BI, BI-UB and FC-UB (n = 89): small_q2_89samples.txt
    • Bootstrapping Results of the Mediation Analysis (Figure 4)
      • Source: SmartPLS 3.0 codes and data (mediation_HM-BI-UB)
      • SAV data: 89_imputed.sav
      • Variable names: variable_names.txt
      • Output in EXCEL files: PLS-SEM Analysis (mediation HM-BI-UB)
      • Note: The Intention-Behavior Model is also referred "Intention-Usage" in the SmartPLS 3.0 codes and data
      • Individual outputs (with the 95% confidence intervals based on on Bias-Corrected and Accelerate (BCa)): Hedonic Motivation (HM) - Behavioral Intention (BI) - Use Behavior (UB)
        • βa = effect of HM-BI
        • βb = effect of BI-UB
        • βc = direct effect HM-UB without mediation
        • βaxb= indirect effect HM-UB (βa x βb)
        • βc' = direct effect HM-UB with mediation (βc - βaxb)