Multivariate Statistical Techniques Research Articles

Product responsibility in the digital age: The impact of the EPR regulation on Hungarian SMEs and the online commerce sector

In this research, we employed multivariate statistical methods to investigate the perspectives of small and medium-sized enterprises (SMEs) concerning the Extended Producer Responsibility (EPR) regulation and their apprehensions related to EPR compliance. The EPR regulation, which places the responsibility of waste management on producers, has significant financial and administrative implications, particularly for SMEs. A sample of 114 businesses was randomly selected, and the collected data underwent comprehensive analysis. Our findings highlight that a notable proportion of businesses (44.7%) possess knowledge of the EPR regulation’s provisions, whereas only a marginal fraction (1.8%) lacks sufficient familiarity. We also explored the interplay between opinions on the EPR regulation and concerns regarding its financial and administrative implications. Our results establish a significant correlation between EPR regulation opinions and concerns, with adverse opinions prominently influencing concerns, particularly regarding financial burdens and administrative workloads. These outcomes, derived from the application of multivariate statistical techniques, provide valuable insights for enhancing the synergy between environmental regulations and business practices. EPR regulation significantly affects SMEs in terms of financial, administrative, and legal obligations, thus our study highlights that policymakers may need to consider additional support mechanisms to alleviate the regulatory burden on SMEs, fostering a more effective and sustainable implementation of the EPR regulation.

Statistical analysis of water quality and cyclone impacts along the south Odisha coast, east coast of India

Water quality along south Odisha coast (25km stretch), east coast of India was studied collecting monthly samples at 10 different sampling stations from June 2008 to March 2011 and June 2012 to May 2017. The samples were analysed for salinity, dissolved oxygen (DO), total suspended solids (TSS) and pH. In addition, river discharge, siltation, rainfall and wind speed data were also collected from 2008 to 2017 at the study region and their impacts on coastal water quality were assessed along with the impacts of cyclonic storms. Anomalies of water quality parameters and their seasonal variability were assessed. Spatio-temporal variation of these parameters and multivariate statistical techniques such as factor analysis, correlation coefficient and, descriptive statistics were applied to the observed datasets. Results indicate that, salinity is positively correlated with wind speed at 0.05 significance level while negatively correlated with river discharge. TSS is positively correlated with siltation at 0.01 significance level. DO values are also positively correlated with strong wind speed. Nutrient driven primary production helps pH to be higher at Rushikulya river mouth during monsoon and post-monsoon as compared to at all other stations. Throughout the year, the study environment remains alkaline. Due to impacts of cyclonic storms, water quality parameters show significant reduction in salinity and enhancement in TSS, DO and pH during the post-cyclone period while the seasonal trend is maintained during the pre-cyclone period. Factor analysis reveals the factors controlling the hydrographic behaviour of seawater. The spatial variation in coastal water quality is very high and dominates over the temporal variation. Importance of coastal water quality assessment in relation to coastal ecosystem and disaster is discussed.

Cluster analysis applied to fuel price data in Campina Grande

This work uses cluster analysis to analyze similarities in the prices of regular gasoline and ethanol in Campina Grande-PB, Brazil. This multivariate statistical technique groups similar elements into distinct clusters using the non-hierarchical method (k-means). The analysis reports three clusters: cluster 1 has the lowest prices, while cluster 3 has the highest. We report descriptive statistics and a discussion based on the empirical local context.

Analysing the Financial Ratios of the Companies in the Maritime Industry with Multivariate Statistical Techniques

This study aims to glance at the shipping sector in terms of financial ratios. To this end, 23 financial ratios were collected from each of the 223 companies worldwide and examined by factor analysis. Seven factors were spotted explaining 85% of the total variance. Subsequently, multidimensional scaling was conducted to analyse the sector on these seven factors. The findings demonstrate profitability ratios as the biggest source of variability and as the reason for the differences among companies. The findings also show that the profitability ratios can be used for monitoring, early warning or decision making purposes for the shipping sector.

Integrated STEM Education Competence Framework for University Lecturers

The rapid advancement of science, engineering, and technology, driven by the Fourth Industrial Revolution, has heightened the demand for a highly skilled workforce in science, technology, engineering, and mathematics (STEM) fields. Integrated STEM education has emerged as a key driver of educational innovation in Vietnam, spanning both general and higher education. The competence of university lecturers in delivering integrated STEM education, a newly recognized pedagogical and professional skill set, is crucial to the success of STEM education at the tertiary level. As with general pedagogical competence, the development of an integrated STEM education competence framework is essential for enhancing this capability among university lecturers. However, there remains a lack of theoretical foundation and best practices tailored to the Vietnamese higher education context. This study aims to develop a framework for integrated STEM education competence specifically for university lecturers through document analysis and survey research. Multivariate statistical techniques, including exploratory factor analysis (EFA), Cronbach’s alpha, and Pearson correlation, were applied to data collected from 205 lecturers across nine public universities in Vietnam. The integrated STEM education competence framework for Vietnamese university lecturers consists of three component competencies and 23 items: designing and implementing integrated STEM education (15 items), assessing integrated STEM learning outcomes (4 items), and demonstrating positive attitudes towards integrated STEM education (4 items). The framework was found to be both reliable and valid, with strong positive correlations among the three component competencies. This study also outlines limitations and provides recommendations for future research.

Application of multivariate statistical techniques in the assessment of long-term surface water quality in Dong Thap Province, Vietnam.

This study assessed the surface water quality in Dong Thap province during 2013-2023 using multivariate statistical techniques. Water quality index (WQI), Pearson correlation, principal component analysis (PCA), and cluster analysis (CA) were determined based on monitoring data for parameters. The findings showed that only some water samples had pH values below the permissible standard, while 95.5% of samples exceeded the standard of Escherichia coli in surface water. In some years, 100% samples having BOD5, TSS, NH4+, and E. coli exceeded the permissible limits. The PCA identified four, five, and seven major potential sources during 2013-2015, 2016-2020, and 2021-2023, respectively, explaining 75.9%, 63.0%, and 63.0% of variation in the corresponding duration times. For heavy metals, the number of water samples containing As, Cu, and Pb exceeding the water permissible limits were 15.5%, 21.4%, and 2.8% samples in 2021, respectively, but no heavy metal was above standard in 2022 and 2023. The evaluation of average water quality based on WQI showed that 79.31% and 76.19% of the sample sites were classified to be average during 2016-2020 and 2021-2023, respectively, and others were bad water quality. However, the water quality showed an improving trend with 70% of monitoring positions considered as good water quality in 2023. The results showed that anthropogenic activities and hydrological regime were the main source of pollution. This study provides guidance to policymakers in developing water management strategies to have proper actions to prevent water pollution.

Evaluation of the dynamics of nutrients and potentially toxic elements along a major river in Ethiopia using multivariate statistical techniques: Implications of possible co-occurrences

This study utilized established field and laboratory methods and multivariate statistical tools to evaluate the nutrient and potentially toxic elements (PTEs) distribution, identify possible sources, and determine potential co-existence in the Awash River. The levels of soluble reactive phosphorus (SRP), total phosphorus (TP), NO3-N, Mn, Ni, Cr, and Cu generally increased downstream in the upper Awash, with the highest respective values of 0.70, 1.34, 1.19, 1.58, 1.03, 0.85, and 0.11 mg L−1 at the inlet to Lake Koka. Sites found downstream of the inflow of Lake Beseka and Metehara town showed the highest levels of Fe, B, Zn, and dissolved silica. Principal components and correlation analysis findings revealed strong associations (r ≥ 0.80, p < 0.05) between nutrient parameters (SRP, TP, and NO3-N) and PTEs (Cu, Ni, Mn, and Cr), implying common origins and potential co-occurrences. The potential co-occurrences of these parameters may strengthen their potential individual impacts due to complex interactions.

Investigation of Ginseng-Ophiopogon Injection on Enhancing Physical Function by Pharmacogenomics and Metabolomics Evaluation.

Ginseng-ophiopogon injection (GOI) is a clinically commonly used drug for Qi deficiency syndrome characterized by decreased physical function in China. This study aimed to clarify common pharmacological mechanisms of GOI in enhancing physical function. We performed an integrative strategy of weight-loaded swimming tests in cold water (5.5°C), hepatic glycogen and superoxide dismutase (SOD) detections, GC-TOF/MS-based metabolomics, multivariate statistical techniques, network pharmacology of known targets and constituents, and KEGG pathway analysis of GOI. Compared with the control group, GOI showed significant increases in the weightloaded swimming time, hepatic levels of glycogen and SOD. Additionally, 34 significantly differential serum metabolites referred to glycolysis, gluconeogenesis and arginine biosynthesis were affected by GOI. The target collection revealed 98 metabolic targets and 50 experimentreported drug targets of ingredients in GOI involved in enhancing physical function. Further, the PPI network analysis revealed that 8 ingredients of GOI, such as ginsenoside Re, ginsenoside Rf, ginsenoside Rg1, and notoginsenoside R1, were well-associated with 48 hub targets, which had good ability in enhancing physical function. Meanwhile, nine hub proteins, such as SOD, mechanistic target of Rapamycin (mTOR), and nitric oxide synthases, were confirmed to be affected by GOI. Finally, 98 enriched KEGG pathways (P<0.01 and FDR<0.001) of GOI were obtained from 48 hub targets of the PPI network. Among them, pathways in cancer, Chagas disease, lipid and atherosclerosis, and PI3K-Akt signaling pathway ranked top four. This study provided an integrative and efficient approach to understand the molecular mechanism of GOI in enhancing physical function.

Hydrochemical characteristics of groundwater and their controlling mechanisms in irrigation and non-irrigation areas - a comparative study in the Guanzhong Plain of China

Gaining insight into groundwater chemistry and its evolution in irrigation areas is essential for optimizing irrigation strategies. This study explores the origins and influencing variables of groundwater chemistry across both irrigated and non-irrigated areas of the Guanzhong Plain, aiming to elucidate how irrigation practices impact hydrochemical changes. Hydrochemical data, isotopic data, multivariate statistical techniques and hydrochemical methods were used in this study. The findings underscored that groundwater in non-irrigation areas had low total dissolved solids, with 90% of the groundwater samples exhibiting the HCO3-Ca•Mg type. In comparison, groundwater from irrigation areas showed elevated total dissolved solids levels and diverse hydrochemical types, including HCO3-Ca•Mg, HCO3-Na, and SO4•Cl-Ca•Mg types. These categories collectively represented 88% of the water samples from irrigation areas. Isotopic and hydrochemical analyses revealed that groundwater in irrigation areas experienced more intense evaporation and cation exchange. Furthermore, anthropogenic activities like sewage and manure had a more pronounced effect on groundwater chemistry in irrigation areas. Groundwater suitability for irrigation deteriorated from the western irrigation area to the eastern irrigation area. This research offers a robust scientific foundation for the careful management and conservation of groundwater resources in the region.

Groundwater hydrogeochemistry and potential health risk assessment through exposure to elevated arsenic in Darrang district of north bank plain of the Brahmaputra

The current study investigates groundwater contamination in Darrang district, situated in the flood-prone Brahmaputra Valley. This research evaluates the concentrations and geospatial distributions of iron, fluoride, and arsenic in groundwater samples (n = 347) and assesses their potential ecotoxicological risks to human health. Multivariate statistical techniques were used to investigate the sources and the mobilization mechanism of the contaminants in the aquifer system. The Piper and Gibbs plots reveal that silicate weathering, ion exchange, and mineral dissolution are the dominant processes influencing groundwater chemistry. Notably, higher pH levels (max = 8.5, mean = 7.4) are associated with alkaline desorption of arsenic from minerals. Cluster analysis also indicates the occurrence of pH induced carbonate weathering process implying weathering and active dissolution as a governing mechanism for arsenic release. The health risk assessments indicate significant non-carcinogenic and carcinogenic risks from prolonged exposure to arsenic, particularly for children. Spatial distribution maps identify areas most at risk of contamination and related health hazards. This is the first comprehensive analysis of groundwater contaminants in Darrang, providing valuable insights into the region’s water quality issues and supporting future groundwater management efforts in Brahmaputra floodplain region.

Evaluating coupled influences of slope class and land use change on water quality using single and composite indices in an agricultural basin

Understanding the critical source areas (CSAs) of pollution and their key relationships with land use is essential for efficient water quality improvement. Despite much efforts to examine their influences on water quality, the role of topographic characteristics in land use processes acting on water quality in lowland regions remains unclear. Therefore, taking a typical rural lowland catchment-the Stör catchment (in northern Germany) as the study area, the influences of different land use types with varying slopes on sediment, total phosphorus (TP), and total nitrogen (TN) loads, as well as on the relevant CSAs of a single or combined index (Ij) were determined using an integrated approach that applies hydrologic model SWAT3s (Soil and Water Assessment Tool, adapted with three groundwater aquifers) and multivariate statistical techniques. The results indicated: (1) sediment and nutrients have markedly declined from 1990 to 1991 over 2009–2011 to 2018–2019, particularly in summer and in comparatively steeper areas with slopes above 5%; (2) the subbasins in the southwest or northeast of the catchment dominated by larger arable fields with slope > 5% and urban areas were identified as CSAs; (3) the spatial heterogeneity of overall water quality was more explained in summer (75% on average) and particularly by croplands (50% on average); (4) water quality indicators were mostly and positively affected by cropland and pasture with relatively steeper slope (>5%). In particular, cropland with slopes > 5% was among those land use indicators producing the highest values of sediment, TP, TN, and Ij. This study demonstrates that the identification of seasonal CSAs could provide insights into the seasonal targeting of conservation practices. Furthermore, communicating model outputs to cause-effect analysis can help formulate slope-adaptability measures for water quality protection.

Design Expert-Supported Method Development and Validation of Cilostazol in Pharmaceutical Formulation Using High-Performance Liquid Chromatography

Background: Developing a simple HPLC method requires an expansive array of literary evidence and experimental routines to perceive the nature of a drug and eventually determine the specific mobile phase and column to be used for attaining better results. Objective: The study aimed to develop and optimize a new, simplified, robust, and sensitive method for the determination of cilostazol in tablets by high-performance liquid chromatography using a Box Behnken design. Methods: The chromatographic separation was carried out on an ODS C18 (4.6 X 250mm and 5μm) column with acetonitrile and methanol (25:75 % v/v) at an effluent flow rate of 1 mL/min and detected at 257 nm. Result: The method was found to be linear in the concentration range of 10-50 μg/mL, and the correlation coefficient was found to be 0.988, and the recovery of cilostazol was 98.16%. The optimized method validated as per ICH Q2A guidelines was found to be accurate, precise, robust, and stable. Conclusion: This research thus throws light on the implementation of statistical multivariate analysis techniques used for drug analysis.

Evaluation of surface water quality in Brahmani River Basin, Odisha (India), for drinking purposes using GIS-based WQIs, multivariate statistical techniques and semi-variogram models

Evaluation of surface water quality in Brahmani River Basin, Odisha (India), for drinking purposes using GIS-based WQIs, multivariate statistical techniques and semi-variogram models

AI-generated imagery in sustainable gastronomy tourism: A study from bottom-up to top-down processing

This study utilized image-based experimental methods to explore the intersection of sustainability, gastronomy tourism, and Artificial Intelligence (AI), with a focus on promoting ‘climate-conscious food’. This term refers to the food produced through eco-friendly practices. Despite the benefits of climate-conscious food for individuals, the environment, and local tourism, its adoption remains limited. Conceptualized on visual information processing, this research comprises two studies employing bottom-up and top-down processing. It confirms that the type of content in AI-generated images, specifically “Food Presentation” versus “Preparation Process,” significantly influences consumers' perceptions of food quality and their purchasing intentions. This effect is moderated by consumers' knowledge of climate-conscious food and their level of environmental concern. By providing empirical evidence analyzed through multivariate statistical techniques, this study offers valuable practical and theoretical insights into how AI can promote climate-conscious food, thereby advancing sustainable gastronomy tourism.

Looking beyond household income status: exploring the interplay of income and other drivers of neighbourhood satisfaction in the Gauteng Province of South Africa

Historic inequalities among different income groups remain a contentious issue affecting neighbourhood sustainability, particularly in South Africa. Prior studies frequently focused on the separate influence of income and other variables of neighbourhood satisfaction, failing to thoroughly explore how these variables intersect and interact with one another. The key objective of this study is to examine the interplay of income and other drivers of neighbourhood satisfaction in the Gauteng Province of South Africa to determine the net importance of household income on one’s neighbourhood satisfaction. The study analyses a sample of 8996 adult respondents interviewed in the Quality of Life 6 Survey (2020/21) conducted in the highly urbanised Gauteng province of South Africa. Bivariate and multivariable statistical techniques are employed to estimate the magnitude of the association between household income and neighbourhood satisfaction. The results show a strong association between household income and individuals’ neighbourhood satisfaction. However, neighbourhood satisfaction based on income alone does not offset the adverse roles of other non-monetary factors as individual and contextual factors. It is imperative to comprehensively understand the interactions between income and other drivers of neighbourhood satisfaction to address various disparities and inequities among income groups and create customised interventions to enhance good housing outcomes and neighbourhood satisfaction in Gauteng.

Designing toys for older adults: Integrating learning theories to address needs

ABSTRACT With advancements in science, technology, and healthcare, senior citizens today face extended retirement periods of 15–35 years. However, many experience challenges such as declining physical and mental functions, which often lead to depression and reduced self-esteem. This study aims to explore the design elements and needs of older adult toys by integrating older adult learning theories to promote successful, healthy, and active aging. Using both qualitative and quantitative research methods, we analyzed expert interviews and applied multivariate statistical techniques, including descriptive statistics and independent sample t-tests, to assess the physical, psychological, and lifestyle needs of older adults. Additionally, we examined older adults’ attitudes and behaviors toward toys, leading to the identification of key design criteria and preferences. Based on these insights, we developed guidelines for designing toys that support the aging process and proposed a model to predict future market trends. The findings offer valuable guidance for the toy industry in fostering innovation and sustainable development.

What innovation and technology delivers to internationalization? The case of European and Central Asia regions

Purpose This study aims to analyze the impact of technology and innovation on businesses’ internationalization process as well as the relationship between innovation and technology and internationalization. Design/methodology/approach The study was based on quantitative analysis of secondary data from 30 economies in Europe and Central Asia, extracted from the Enterprise Surveys – The World Bank database, using the multivariate statistical techniques of multiple linear regression and cluster analysis. Findings The results suggest that innovation and technology (I&amp;T) mutually and positively influence the internationalization of firms. In this sense, innovative firms are more successful in competing internationally and exposure to alternative business contexts culminates in innovation. The results suggest a positive relationship between the percentage of firms that introduced a process innovation, the percentage of firms with websites and the businesses’ internationalization. Originality/value The originality of this research is based on the methodological approach that allowed the identification of the most significant variables for stimulating the internationalization of firms based on innovation and technology. Given the predominance of economies with higher I&amp;T and internationalization in Europe, it is proposed as future research projects to ascertain the reasons for such a phenomenon.

Age Structure, Body Size, and Sexual Dimorphism in a High-Altitude Population of Pelophylax ridibundus (Pallas, 1771).

This study aimed to describe the morphological characteristics and age structure of a P. ridibundus population sampled from a high altitude. For this purpose, a total of 54 adult frogs (33 males and 21 females) were collected from Şavşat district, Artvin province, Türkiye. The samples were aged based on skeletochronology method. Furthermore, von Bertalanffy growth curve models were constructed using SVL and weight data. To assess sexual size and shape dimorphism, the measurements of 23 different morphological characters were analyzed using univariate and multivariate statistical techniques. The mean SVL and weight were found to be 78.05 mm and 60.87 g for males and 93.95 mm and 109.61 g for females. The mean age was 3.67 years and 4.05 years for males and females, respectively. The maximum age was found to be 6 years for both sexes. Growth curve models revealed a significant relationship between age, SVL, and weight. A clear female-biased sexual size dimorphism was found in all morphological characters. However, SVL-adjusted values indicated that males had larger head and limb structures resulting from evolutional and sexual pressures associated with reproduction and male-male competition. The shape dimorphism was found only in forelimbs, and it was male-biased and was related to clasping in amplexus.

Assessing groundwater quality in the northwestern coastal area of Sri Lanka using multivariate statistical techniques

The objective of the present study was to assess groundwater quality in the northwestern coastal region of Sri Lanka by analyzing hydrogeochemical data using multivariate techniques. The study examined the relationship between water quality and sources influencing groundwater quality and proposed useful statistical tools for water resource management. Descriptive, comparative, correlation, factor analysis (FA), and discriminant analysis (DA) were among the techniques used. The water quality in terms of electrical conductivity (EC), total dissolved solids (TDS), and dissolved nitrate and phosphate concentrations of the northwestern coastal area, where different aquifer conditions prevailed, were compared, and differences in the mean concentrations of the above parameters were identified. The study also employed ridge regression and support vector machines (SVM) to predict nitrate levels and validate the results using a 70:30 train/test split. The analysis found that the mean concentrations of EC, TDS, and HCO3- are higher in the sandy aquifer area, while Fe2+ and PO43- are higher in the regolith aquifer area. EC, TDS, PO43-, and HCO3- in the hard-rock aquifer area exceed permissible limits. Factor analysis identified four factors contributing to pollution, and a linear discriminant model classified all observations with 99.9% accuracy. The water quality index for Kalpitiya, Vanathavillu, and Katana-Negombo areas was 68.1, 99.2, and 287.9, respectively, while the World Health Organization (WHO) cutoff was 50. The use of multivariate statistical techniques demonstrated the value of analyzing hydrogeochemical data to gain valuable insights into the complex interplay of factors affecting groundwater quality. The SVM model, using a linear kernel, was identified as the best model to predict nitrate levels. A combination of statistical analyses reveals that, with the exclusion of a few wells, the groundwater in the study area is polluted and inappropriate for human consumption or culinary purposes. Nonetheless, the inhabitants persist in utilizing it as their primary source of drinking water. This study reveals the combination of statistical analyses is useful for the management of groundwater resources.

Rill Erosion Due to Wildfire or Deforestation in Forestlands of Northern Iran

Rill erosion, mostly affecting steep and long hillslopes, is one of the most severe effects of deforestation and wildfires in natural ecosystems. Specific monitoring and accurate but simple models are needed to assess the impacts of these forest disturbances on the rill detachment process. To address this need, this study has simulated the rill detachment capacity (Dc) through flume experiments on samples of soils collected in hillslopes after deforestation and severe burning. The associations between Dc and organic matter (OM) and the aggregate stability of soil (WSA), two key parameters influencing the rill detachment process, have also been explored under the two soil conditions (deforested and burned soils) using multivariate statistical techniques. Finally, linear regression models to predict Dc from these soil parameters or the hydraulic and morphological variables (water flow rate, WFR, and soil slope, S), set in the flume experiments, have been proposed for both soil conditions. Higher Dc in samples from deforested sites compared to the burned soils (+35%) was measured. This Dc increase was associated with parallel decreases in OM (−15%) and WSA (−34%) after deforestation compared to the wildfire-affected sites. However, the discrimination in those soil properties between the two soil conditions was not sharp. Accurate linear equations (r2 &gt; 0.76) interpolating Dc and the shear stress (τ) have been set to estimate the rill erodibility (Kr) to evaluate soil resistance in erosion models to be applied in deforested or burned sites.