Open Innovation Capability and Digital Procurement Transformation in Vietnamese SMEs
Description
Data for this study were collected through a structured survey targeting managers and decision-makers responsible for procurement, supply chain management, and digital transformation within Vietnamese small and medium-sized enterprises (SMEs). A purposive sampling approach was employed to ensure that respondents possessed relevant domain knowledge and practical experience in procurement and innovation-related activities. The target population included firms operating in manufacturing, logistics, and retail sectors, where digital procurement transformation is increasingly critical. The questionnaire was initially developed in English based on validated scales from prior literature and subsequently translated into Vietnamese using a back-translation technique to ensure semantic equivalence. To enhance content validity and clarity, a pilot test was conducted with a group of academic experts and industry practitioners. Feedback from the pilot phase was incorporated to refine wording, eliminate ambiguity, and ensure contextual relevance. Data collection was conducted between January and June 2024 using an online survey platform (Google Forms) and through direct distribution via professional networks, business associations, and email invitations. Participation was voluntary, and respondents were assured of anonymity and confidentiality to reduce potential response bias and encourage honest reporting. A total of 468 responses were collected. After data screening procedures, including the removal of incomplete responses, detection of straight-lining patterns, and consistency checks, 420 valid responses were retained for analysis. The final sample size satisfies recommended thresholds for Partial Least Squares Structural Equation Modeling (PLS-SEM) and is also adequate for fuzzy-set Qualitative Comparative Analysis (fsQCA).
Files
Steps to reproduce
Analysis Resources To ensure full transparency and reproducibility, all statistical scripts, analytical procedures, and model specifications used in this study have been systematically documented and made available. The analytical workflow consists of three complementary techniques: Partial Least Squares Structural Equation Modeling (PLS-SEM), fuzzy-set Qualitative Comparative Analysis (fsQCA), and Necessary Condition Analysis (NCA). (1) PLS-SEM Model Estimation PLS-SEM was employed to examine the linear relationships among constructs and to test the proposed hypotheses. The analysis followed a two-stage approach, including measurement model evaluation and structural model assessment. The measurement model was assessed in terms of internal consistency reliability (Cronbach’s alpha and composite reliability), convergent validity (average variance extracted – AVE), and discriminant validity (Heterotrait–Monotrait ratio – HTMT). The structural model was evaluated using path coefficients, bootstrapping procedures, coefficient of determination (R²), effect sizes (f²), and predictive relevance (Q²). All estimation procedures, bootstrapping routines, and model outputs are fully documented in the provided scripts. (2) fsQCA Procedures fsQCA was applied to identify configurational pathways leading to high levels of digital procurement transformation. The analysis involved three main steps: Calibration: All variables were transformed into fuzzy-set membership scores using the direct calibration method based on three qualitative anchors: full membership, crossover point, and full non-membership. Truth Table Construction: A truth table was generated with predefined frequency and consistency thresholds to identify empirically relevant configurations. Solution Derivation: Intermediate and parsimonious solutions were computed to determine sufficient configurations, with consistency and coverage used as evaluation criteria. Robustness checks were conducted by varying calibration thresholds and consistency cut-offs to ensure the stability of results. (3) Necessary Condition Analysis (NCA) Where applicable, NCA was conducted to identify necessary conditions for achieving high procurement performance. The analysis involved calculating effect sizes and bottleneck levels to determine whether specific conditions must be present for the outcome to occur. The NCA procedure complements PLS-SEM and fsQCA by distinguishing necessary from sufficient conditions, thereby enhancing the explanatory power of the study. (4) Model Specifications and Formulas All model specifications and analytical formulas used in this study are fully documented. These include: Structural equations for PLS-SEM; Calibration functions for fuzzy-set transformation; Consistency and coverage metrics for fsQCA; Effect size and necessity metrics for NCA. Detailed scripts and step-by-step procedures are provided in the accompanying repository, allowing full replication of all reported results.
Institutions
- FPT UniversityHanoi, Hanoi