Hypothesis Testing Research Articles

Recent Advancements in Anomaly Detection for Manufacturing Systems

Recent advancements in anomaly detection and root cause analysis have revolutionized manufacturing processes, enabling improved efficiency, quality control, and cost reduction. This review explores the latest developments in these fields, focusing on supervised and unsupervised anomaly detection methods, statistical techniques for root cause analysis, and predictive modeling approaches. Supervised anomaly detection methods, such as classification-based approaches, support vector machines, and decision trees, leverage labeled data to identify anomalies. Unsupervised techniques, including clustering-based methods (e.g., k-means and DBSCAN), density-based approaches (e.g., local outlier factor), and dimensionality reduction (e.g., PCA and autoencoders), detect anomalies without prior knowledge of anomalous instances. Statistical methods for root cause analysis, such as statistical process control, hypothesis testing, correlation analysis, and multivariate techniques (e.g., factor analysis), enable the identification of the underlying causes of manufacturing issues. Predictive modeling techniques, including time-series forecasting, machine learning for predictive maintenance, and optimization methods for process improvement, further enhance manufacturing efficiency. The integration of these techniques has led to significant improvements in quality control, equipment failure prediction, and supply chain optimization. However, challenges remain in handling complex, high-dimensional data and integrating domain knowledge with data-driven approaches. Future research directions include the development of explainable AI, edge computing, and federated learning for anomaly detection in manufacturing. This analysis offers a thorough examination of cutting-edge techniques for detecting anomalies and analyzing root causes, emphasizing their capacity to transform manufacturing operations and stimulate future advancements in these research areas. Keywords—anomaly detection, manufacturing, supervised and unsupervised anomaly detection

A Review of Mendelian Randomization in the Presence of Weak Instrumental Variables; Statistical Methods and Challenges

Introduction: Mendelian randomization (MR) assesses the causal effect of risk factors by using genetic variations as instrumental variables (IV) in nonexperimental data. IV strategies are one of a few available methods for determining causal effects in the absence of specific knowledge of all confounders in the exposure- outcome relationship. To use an IV as a legitimate instrumental variable, it must meet the following criteria: relevance, exchangeability, and exclusion restriction. A weak instrument is a circumstance in which there is a piece of weak statistical evidence for an association between IV and exposure. Weak instruments cause significant issues, including (i) insufficient statistical power to hypothesis testing, (ii) increasing bias with deviation from IV assumptions, and (iii) asymptotic estimation of standard errors and confidence intervals. Therefore, in this study, we intend to introduce the Mendelian randomization method, weak instrumental bias, and statistical remedy methods used in this bias. Methods: Current study was conducted by using Medline/PubMed, Scopus, Web of Sciences and Google Scholar. Results: This review provides a comprehensive description of the principles of MR, and a guide to basic MR methodology. To deal with these challenges, the bulk of the review considered statistical remedies. The review ends with a section that details the practical limitations, and recommendations regarding MR and the weak instruments Conclusion: Depending on the type of data, several solutions can be used in one and multiple IV. Moreover, it can be used in solutions in the design and analysis phase to minimize the effects of weak instruments

Exploring the Influence of Brand Authenticity on Consumer Behavior: Insights from Generation Y Consumers in Pakistan

This study investigates the influence of perceived brand authenticity (PBA) dimensions (quality commitment, heritage, and sincerity) on brand love and subsequent consumer behavior among Generation Y consumers in Pakistan. Additionally, this research introduces brand trust as a mediating variable and explores differences between local and global brands to enhance originality. A mixed-approach approach emerges as employed, combining quantitative surveys and qualitative interviews. Data were accumulated from a numerous sample of Generation Y clients in Pakistan through a hooked up questionnaire. Partial Least Squares Structural Equation Modeling (PLS-SEM) emerges as used for hypothesis testing. Additionally, in-depth interviews were achieved to capture qualitative insights on brand authenticity perceptions. The results reveal that PBA dimensions significantly impact brand love, which, in turn, influences continuous purchase intention (CPI) and price premium (PP) willingness. Moreover, brand trust mediates the relationship between PBA and consumer behavior, strengthening the effect. Additionally, local brands (e.g., J., Khaadi, Daraz.pk) exhibit stronger authenticity perceptions than global brands (Adidas, KFC, Amazon) among Pakistani consumers. Digital marketing and influencer endorsements further enhance authenticity perceptions. This study contributes to branding literature by integrating Self-Determination Theory (SDT) and Consumer Culture Theory (CCT) to explain Pakistani consumers’ preference for authentic brands. It highlights the role of trust and digital branding in shaping PBA perceptions. For marketers, the findings suggest that authenticity-driven strategies, including transparent communication, cultural storytelling, and social media engagement, can enhance brand loyalty and justify premium pricing. This research contributes to the existing literature by examining the role of PBA in shaping consumer behavior within the Pakistani context. By understanding the specific factors that drive brand love and loyalty among Generation Y consumers, marketers can develop more effective branding strategies.

Exploring the Correlation of Physiological Stress Signals with Student Exam Performance: A Preliminary Study.

Stress responses in real-world settings are less studied compared to controlled laboratory environments, limiting our understanding of their impact on cognitive performance. This study investigates the relationship between physiological stress signals and academic performance using an open-access dataset of 10 students assessed across three exam sessions (Midterm 1, Midterm 2, and Final Exam). Physiological measures, including electrodermal activity (EDA), heart rate (HR), and skin surface temperature (TEMP), along with exam grades, were analyzed using traditional hypothesis testing, bootstrap method, correlation analysis, and regression tree modeling. To address the small sample size, we validated traditional hypothesis test results with the bootstrap method, and both approaches were compatible. Hypothesis testing revealed no significant differences in physiological measures across exam sessions, supporting the null hypothesis. Grades differed significantly between the final exam and both midterms (p < 0.05). Stress fluctuations were also analyzed across three periods (beginning, middle, and end) for each exam, revealing temporal response variations. Correlation analysis showed a moderate negative relationship between EDA and HR (r = -0.504, p < 0.01) and a weak positive relationship between EDA and TEMP (r = 0.417, p < 0.05), both intensifying during the final exam. Regression analysis explained 78% of the variance in grades (R2 = 0.78), with regression tree modeling identifying lower skin temperature (< 28°C) and higher EDA (≥ 0.19) as predictors of poorer performance. These findings underscore the interplay between physiological stress responses and academic outcomes, emphasizing the need for further research and interventions to support student success.

Construct validity, responsiveness, and interpretability of the Spinal Cord Injury Functional Ambulation Inventory (SCI-FAI)

Objective We investigated the construct validity, responsiveness, and interpretability of the Spinal Cord Injury Functional Ambulation Inventory (SCI-FAI) to determine its usefulness in measuring the functional level of gait. Patients and Methods This was a prospective observational study following the checklist of the Consensus-Based Standards for Selecting Health Measurement Instruments. The SCI-FAI consists of three items: Gait Parameter, Assistive Devices, and Temporal. We recruited 55 patients diagnosed with a spinal cord injury (SCI) in the subacute or chronic phase from Ibaraki Prefectural University of Health Sciences Hospital. Construct validity was clarified by hypothesis testing and the rate of hypothesis verification. We divided responsiveness into subacute and chronic groups and determined the correlation between the changes in the two timepoints of the SCI-FAI and 6-minute walk test. Interpretability involved the calculation of cutoff values for indoor and outdoor walking. Results Construct validity was confirmed for 12 of the 15 hypotheses, indicating high construct validity. In the subacute group, Gait Parameter and Assistive Devices showed moderate responsiveness. Interpretability showed that the Gait Parameter was perfect in 19 of the 20 subjects who did not need a walking aid when walking. The cutoff value for Gait Parameter for indoor walking was 17.5 points (AUC 0.91) and that for Assistive Devices was 9.5 points (AUC 0.88). The cutoff values for outdoor walking were Gait Parameter 18.5 points (AUC 0.96) and Assistive Devices 10.5 points (AUC 0.94). Conclusion Our results demonstrated that the SCI-FAI has adequate construct validity, moderate responsiveness in SCI patients in the subacute phase, and interpretability in the gait assessment of individuals with SCIs. Gait Parameter is likely to show a ceiling effect for people with SCIs who can walk without using an upper-limb walking aid.

The Influence of Small-Scale Businesses in Employment Generation and Economic Growth in Nigeria: A Case Study of Selected Business in Suburban Lagos

This research examines the roles of small-scale business (SSBs) in employment generation and financial growth, with a focus on businesses within the Ikorodu Local Government Area. The study employs a systematic methodology to address the research objectives, including hypotheses testing, population and sample selection, data collection, and analysis. A sample of fifty (50) respondents, drawn using simple random and stratified sampling techniques, provides a fair representation of the study population. Primary data were collected through questionnaires, ensuring cost-effective, time-efficient, and accurate data acquisition. The collected data were analyzed using frequency distribution tables and the chi-square test to assess whether observed proportions align with expected values. The findings reveal insights into the demographic and business characteristics of the sample, including gender distribution, age range, business duration, and educational qualifications of the respondents. The chi-square analysis tested two hypotheses: the significant role of SSBs in employment generation and financial growth. Results indicate that small-scale business significantly contribute to employment generation and the financial growth of the economy. This research offers valuable insights into the potential of small-scale businesses as drivers of economic development, with implications for policy and practice in promoting sustainable entrepreneurship.

Neuroeconomics and Neuroenvironment in Chimborazo Markets: Post-Pandemic Consumer Behavior

Objective: The study aimed to analyze the influence of neuroeconomics and neuroenvironment on consumer behavior in the markets of Guano, Ecuador, during the post-pandemic period. The primary focus was on understanding how these factors affect purchasing decisions in local gastronomic businesses, particularly in the context of the economic and social changes brought about by the COVID-19 pandemic in the region. Method: Method: The research employed a mixed-methods approach combining quantitative and qualitative methodologies. Using a finite population formula, a sample size of 382 consumers was calculated from a total population of 42,851 residents. The study utilized advanced eye-tracking technology to measure visual attention and consumer behavior, complemented by in-depth interviews and focus groups. Data collection was conducted through simple random sampling, and statistical analysis was performed using SPSS software, with Chi-squared tests employed for hypothesis testing to evaluate the relationship between neuroeconomic factors and consumer behavior. Results and Discussion: The analysis revealed a significant correlation between neuroeconomics, neuroenvironment, and consumer behavior, with a statistical significance of p-value = 0.000. The study identified key emotional responses including joy, trust, and curiosity, with biosafety and cleanliness emerging as primary factors in building consumer trust. Customer service quality and appropriate ambient music generated positive emotional responses, while eye-tracking data demonstrated varying levels of consumer engagement with brands. Interestingly, certain aspects of market decoration were found to induce negative emotions such as sadness. These findings conclusively validated the hypothesis that neuroeconomics and neuroenvironment significantly influence purchasing decisions in Guanese markets. Research Implications: The study's findings provide crucial insights for traders and marketers to enhance consumer experiences in post-pandemic market environments. The research emphasizes the importance of prioritizing biosafety and cleanliness measures, maintaining product variety to stimulate consumer curiosity, and creating engaging market environments. These insights offer valuable guidance for developing effective marketing strategies adapted to post-pandemic consumer behavior, particularly beneficial for local businesses adapting to new market dynamics and consumer expectations. Originality/Value: This research pioneers the examination of post-pandemic consumer behavior in Chimborazo markets using neuroeconomics and eye-tracking technology, offering valuable insights for local businesses adapting to new market dynamics and contributing to the understanding of consumer behavior during crisis periods.

Parameter Estimation and Hypothesis Testing of Truncated Normal Uncertainty Distribution for Non-Negative Variables

Parameter Estimation and Hypothesis Testing of Truncated Normal Uncertainty Distribution for Non-Negative Variables

A dataset for evaluating clinical research claims in large language models

Large language models (LLMs) have the potential to enhance the verification of health claims. However, issues with hallucination and comprehension of logical statements require these models to be closely scrutinized in healthcare applications. We introduce CliniFact, a scientific claim dataset created from hypothesis testing results in clinical research, covering 992 unique interventions for 22 disease categories. The dataset used study arms and interventions, primary outcome measures, and results from clinical trials to derive and label clinical research claims. These claims were then linked to supporting information describing clinical trial results in scientific publications. CliniFact contains 1,970 instances from 992 unique clinical trials related to 1,540 unique publications. When evaluating LLMs against CliniFact, discriminative models, such as BioBERT with an accuracy of 80.2%, outperformed generative counterparts, such as Llama3-70B, which reached 53.6% accuracy (p-value < 0.001). Our results demonstrate the potential of CliniFact as a benchmark for evaluating LLM performance in clinical research claim verification.

What is the experimentalist challenge to the method of cases?

This paper puts forward a new formulation of the experimentalist challenge to the method of cases. Unlike previous attempts to articulate the challenge, the one proposed is based on a clear characterization of the targeted philosophical methodology. The method of cases is explicated as a form of thought experimentation aimed at testing philosophical hypotheses with a distinctive modal force. Given this explication, the empirical evidence gathered by experimental philosophers concerning the instability of case judgments is shown to constitute a stumbling block for the validation of the method as a rational form of hypothesis testing. The experimental challenge is presented as the challenge to compellingly indicate how this stumbling block can be removed, and the method of cases validated as a form of thought experimentation suitable for rationally testing philosophical hypotheses. The challenge, as presented in the text, is shown to be free from all the main criticisms leveled at previous versions of it and as one not yet met.

Influence of Financial Management Accountability and The Spirituality of Pastoral Services In the Parishes of The Tomohon Victory Towards The Achievement of Pastoral Work

The Manado Diocese Catholic Church implements a Church financial management system within the framework of transparent, objective and responsible financial management. In this context, it is necessary to involve lay pastoral staff who are considered capable and competent and have a high spirituality of service in order to realize the pastoral works of the Manado Diocese Church. This research aims to examine the phenomenon of the influence of accountability in parish financial management and the spirituality of pastoral servants in the context of achieving pastoral work. This research was conducted in all churches or parishes in the Kevikepan Tomohon area by taking 95 samples from 133 populations. The research method used is a quantitative research method with a survey approach. Through this method, researchers want to test hypotheses and at the same time explain the relationship between variables. The data analysis process uses the application of correlation statistics and multiple regression test to analyze the influence between variables and measure the level of relationship between each variable studied. The research results show that: First, there is a significant and linear influence between Financial Management Accountability and Pastoral Work Achievement, meaning that if Financial Management Accountability is implemented well and appropriately it will increase Pastoral Work Achievement in the Tomohon Kevikepan Parishes. Second, there is a relationship between the Spirituality of Pastoral Servants and the Achievement of Pastoral Work. This means that if the Spirituality of Pastoral Servants is interpreted well and effectively, then this will have the effect of increasing the Achievement of Pastoral Work in Tomohon Kevikepan Parishes. Third, the results of hypothesis testing and multiple correlation and multiple regression analysis prove that there is an influence between Financial Management Accountability and Pastoral Servant Spirituality together on Pastoral Work Achievement. Furthermore, it can be concluded that an accountable financial management model supported by the spirituality of pastoral servants can increase the achievement of pastoral work in parishes.

Building Trust through ESG: A Study on Consumer Behaviour in Sustainable Fashion

Background This study explores the influence of Environmental, Social, and Governance (ESG) practices on consumer purchasing decisions in the fashion industry. It examines how corporate image and word-of-mouth (WOM) mediate the relationship between ESG practices and consumer trust, impacting purchase intent. Methods A structured questionnaire was distributed online to 393 participants aware of ESG practices, targeting Generation Z consumers. The survey assessed demographics, ESG awareness, brand image, WOM, and purchasing decisions using a Likert scale across five constructs: ESG Practices, Corporate Image, WOM, Trust, and Purchase Intention. Data were analyzed with Structural Equation Modeling (SEM) in IBM AMOS, following a two-step process: Confirmatory Factor Analysis (CFA) for validation and path analysis for hypothesis testing. Results The findings indicate that ESG practices significantly enhance corporate image and positive WOM, which, in turn, foster consumer trust. Corporate image has a substantial impact on consumer trust (p &lt; .001), while WOM also contributes to trust (p &lt; .001). Trust was found to be a strong predictor of purchase intention, demonstrating the mediating role of corporate image and WOM between ESG practices and purchasing behavior. Conclusions ESG practices are critical in shaping consumer perceptions and fostering loyalty in the fashion sector. By building a trustworthy corporate image and encouraging positive WOM, fashion brands can effectively influence consumer purchasing decisions. This study highlights the competitive advantage of implementing ESG practices, particularly in appealing to Generation Z consumers who value sustainability and ethical business practices.

Testing hypotheses of skull function with comparative finite element analysis: three methods reveal contrasting results.

Comparative finite element analysis involves standardising aspects of models to test equivalent loading scenarios across species. However, regarding feeding biomechanics of the vertebrate skull, what is considered "equivalent" can depend on the hypothesis. Using 13 diversely-shaped skulls of marsupial bettongs and potoroos (Potoroidae), we demonstrate that scaling muscle forces to standardise specific aspects of biting mechanics can produce clearly opposing comparisons of stress or strain that are differentially suited to address specific kinds of hypotheses. We therefore propose three categories of hypotheses for skull biting mechanics, each involving a unique method of muscle scaling to produce meaningful results: those comparing (1) the skull's efficiency in distributing muscle forces to the biting teeth, via standardising input muscle force to skull size, (2) structural biting adaptation through standardising mechanical advantage to simulate size-independent, equivalent bites, and (3) feeding ecology affected by size, such as niche partitioning, via standardising bite reaction force.

The impact of perceived usefulness, ease of use, trust, and usage attitude on the intention to maintain engagement in AR/VR sports: An exploration of the technology acceptance framework

This study examines how perceived usefulness, ease of use, trust, and usage attitude influence the intention to maintain engagement in AR/VR sports. Virtual Reality (VR) and Augmented Reality (AR) technologies have garnered significant attention in the sports domain in recent years, holding promise for substantial economic and social benefits in the future. However, factors influencing user acceptance and continued use of AR/VR remain unclear. To foster sustained adoption of AR/VR sports technologies and meet the growing demand for diverse forms of athletic activities, this study investigates the factors influencing Chinese sports university students' willingness to continue participating in VR/AR sports using the Technology Acceptance Model (TAM). Data were collected through a questionnaire survey among 283 Chinese sports major university students who had participated in VR/AR sports. Data processing and hypothesis testing were conducted using SPSS 26.0 and AMOS 24.0. The main findings of this study are as follows: First, perceived usefulness has a positive influence on usage attitudes. Second, perceived ease of use positively influences user attitudes. Third, trust positively influences usage attitudes. Fourth, usage attitudes have a positive influence on behavioral intentions. These findings expand the application and explanatory power of TAM in the field of AR/VR sports and offer practical implications for the industry. They can serve as a reference for developing marketing strategies to encourage continued consumer use of these services in the future.

KINERJA KARYAWAN PRODUKSI CV.XYZ DIPENGARUHI OLEH DISIPLIN KERJA DAN BUDAYA ORGANISASI

Every organization must strive to achieve company goals. One aspect that can support progress within a company is employee performance. This finding intends to see how much the contribution of work discipline and cultural organizational culture to employee performance. Distributed questionnaires in this study were distributed to 40 production employees who were used as samples. The data was then processed using Smart Partial Least Square (PLS) to determine the validity, reliability, and hypothesis tests. This study shows the results that employee performance is positively influenced by work discipline and organizational culture. Employees who are able to obey all company regulations and are able to apply the culture set by the company will make employees able to improve their performance.

KAJIAN FRAUD HEXAGON DALAM MENDETEKSI FRAUDULENT FINANCIAL REPORTING PADA PERUSAHAAN MANUFAKTUR SEKTOR MAKANAN DAN MINUMAN DI BEI (2020-2022)

This study aims to determine the effect of fraud hexagon on fraudulent financial reporting. The sample of this study were 52 food and beverage sector manufacturing companies listed on the Indonesia Stock Exchange during the period 2020 - 2022. The sampling technique used was non-problability sampling technique using purposive sampling method and data testing was carried out using SPSS 25. The data analysis techniques used include descriptive statistical analysis, pooling test, classical assumption test, and hypothesis testing. The results showed that the total accrual ratio had a positive effect on fraudulent financial reporting, while financial targets, change in director, political connection, ineffective monitoring and CEO duality had no effect on fraudulent financial reporting.

The Effect of Earning Management, Information Asymmetry and Growth Asset on Future Stock Return with Dividend Policy as the Moderating Variable

This study aims to test and analyse the effect of effect of Eamings Management, Information Asymmetry and Asset Growth on the Future Stock Returns and whether the Dividend Policy could strengthen the relationship between the effects of Eamings Management, Information Asymmetry and Asset Growth on the Future Stock Returns in index LQ-45 companies listed on the Indonesia Stock Exchange from 2017 to 2022. This type of research is quantitative, using secondary data. The hypothesis testing used in the study was the Moderated Regression Analysis and Eviews 13 applications. The population in this study were all companies listed on index LQ-45. The data collection technique in this study was purposive sampling technique, from 45 research populations to 18 used as research samples. The results showed that Eamings Management, Information Asymmetry and Asset Growth simultaneously affect Future Stock Returns in index LQ-45 companies from 2017 to 2022. The research results obtained Risk Eamings Management affects Future Stock Returns, Information Asymmetry has no effect on Future Stock Returns, Asset Growth has no effect Future Stock Returns and the dividend policy could not strengthen the relationship between the effect of eamings management, information asymmetry and asset growth on the future stock retums.

Refining the network for the Chinese analytic causative constructicon: insights from corpus analysis and psycholinguistic experimentation

Abstract This study explores the network of the Chinese analytic causative constructicon adopting a cross-variety perspective. Unlike prior research focusing solely on the causative marker, this study acknowledges the multi-level nature of the constructicon and examines the descriptive causative construction identified by Zhang, Lili. (2005. Cong shiyi dao zhishi [The causativisation of Chinese shi-yi constructions]. Taida Wenshi Zhexue Bao (62). 119–152), which features a syntactically specified causee slot and a semantically defined effect predicate slot, distinguishing it from general causatives. Combining corpus analysis and psycholinguistic experimentation, our research reveals a consistent preference for using lìng as the causative marker within the descriptive causative construction, observed in both Mainland and Taiwan Chinese. By analyzing the syntactic and semantic features of the descriptive causative construction and its association with the causative marker lìng, we argue that this construction is stored at a separate level. Furthermore, we propose that the Chinese analytic causative constructicon is best represented by a radial network, accommodating both hierarchical and associative relationships, rather than a purely hierarchical structure as illustrated in Pijpops, Dirk, Dirk Speelman, Freek Van De Velde &amp; Stefan Grondelaers. (2021. Incorporating the multi-level nature of the constructicon into hypothesis testing. Cognitive Linguistics 32(3). 487–528).

Are household facilities influencing biomarkers? Evidence from the LASI survey, 2017–18

In light of India's rapidly ageing population, projected to make up 40% of the population by 2050, this study uses information from the Longitudinal Ageing Study in India (LASI) to examine how Household Environment Condition (HEC) affects the health of older adults. By concentrating on biomarkers, the research seeks to understand how these health indicators are spread across different states and their connections to household facilities. The results are expected to offer detailed viewpoints for policy actions, helping improve living conditions and health for the elderly in India.The aim of this study is to explore the access of household facilities of older people from different states in India and to check the association between the available household facilities and biomarkers.This study used data from the Longitudinal Ageing Survey of India (LASI), executed in 2017-18 where 24,323 individuals with biomarkers specifically collected from those aged 60 years and above. The analysis, employs various statistical techniques, including descriptive analysis, correlation analysis, and hypothesis testing, providing a robust methodology to explore the intricate relationships between biomarkers and household facilities among the elderly population in India.Significant variations were observed in body mass index, systolic blood pressure, waist-to-hip ratio, and grip strength between the male and female cohorts. Positive associations were observed between housing and cooking fuel, housing and a separate kitchen, and cooking fuel and a separate kitchen. However, a negative correlation was noted between housing and safe water, suggesting a potential trade-off. Significant correlations between household amenities and health biomarkers suggest a potential influence of living conditions on health outcomes in the elderly citizens of India. This study highlights diverse demographic and health characteristics, emphasizing the urgent need for interventions addressing educational gaps, economic disparities, and health inequalities among the elderly in India for overall well-being of ageing population.

Development of Sme Based Opinion Leadership Model in Indonesia

The decline in sales is one of the impacts of the Covid 19 pandemic on SME, and another impact is a decrease in production and a decrease in the total income of SME. Opinion leadership can solve consumer behavior in Indonesian society, especially after the Covid 19 pandemic. This research aim to determine the SME-based opinion leadership model in Indonesia during the Covid 19 pandemic and to analyze the relationship between opinion leadership and consumer behavior in SMEs during the Covid 19 pandemic. This research is quantitative through an online survey, the data analysis and hypothesis testing method use the Structural Equation Model - Partial Least Square (SEM-PLS). It was found that opinion leadership significantly affects consumer behavior in SMEs in Indonesia with four characteristics, professional, innovative, involved, and socialized. This research can be used for SME to start digitalizing by having a social media account to share information alongside the efforts of opinion leaders. Opinion leaders have a specialty depending on their capability, so they can promote SME products that match their personality. In addition, having many followers can make opinion leaders lead a community.