Step Cluster Analysis Research Articles

Polish consumer food choices and beliefs about organic food

PurposeThe purpose of this paper is to examine consumers' motives for food choice and to reveal beliefs about organic food.Design/methodology/approachThe paper is based on results of a survey conducted on a sample of 1,010 Polish consumers. The questionnaire included items to identify food choice motives, attributes of organic food and barriers to buy organic products. The data collected was analyzed with SPSS 15.0 using descriptive statistics and carrying out two step cluster analysis to identify consumers' segments. To get better insight into factors decisive to organic food consumption, the segments were profiled according to declared consumption of organic food.FindingsSensory factors were the most important motives for food choice, followed by price and safety. Consumers with convenience and price orientation in their food choices were less inclined to buy organic products while consumers open to novelties and willing to try new foods more often declared to buy organic products. Polish consumers conceptualize organic food referring to aspects such as healthiness and safety. Despite the developments observed in the organic food sector in Poland, the information related that barriers to buy organic food still prevail.Research limitations/implicationsThe choice of validated scale items related to food choice and food related lifestyle was limited and there is a need to further explore Polish consumers food related lifestyle and its relation to organic food consumption. Cluster analysis used to identify segments is a method not based on probabilistic statistics, so it is the researcher's interpretation that is most relevant.Originality/valuePrior research concerning Polish consumers food choices with relation to organic food consumption is very limited. Such information is particularly pertinent to food manufacturers and distributors to support developing communication strategies to stimulate organic food market development in Poland.

Learners’ Learning Style Classification related to IQ and Stress based on EEG

The importance to recognize a learner's Learning Style (LS) is ever-essential as to substantiate success in a teaching and learning process. At the same time, the learner's IQ and personality traits such as Stress also being actively investigated in educational research as educationists consistently attempted to understand learners in a more adept way. Nevertheless, the effort was usually confined to psychoanalysis test. With the emergence of Electroencephalography (EEG) technology, learner's brain characteristics could be accessed directly and the outcome may well hand-in-hand supported the conventional test. In this study, the participants (n= 80) are grouped to the LS of Diverger, Assimilator, Converger or Accommodator using the Kolb's Learning Style Inventory (KLSI). Subsequently, their brain signals were then recorded using EEG at resting baseline state of Open Eyes and Closed Eyes. A statistical tool of SPSS 16 was used for data analysis purposes. Using the Two Step Cluster analysis, the participants’ EEG datasets were 100% classified to the corresponding LS. Then, EEG Alpha band was selected to link between LS, IQ and Stress. The study concluded that Diverger is the LS with highest IQ while Converger and Diverger are the LS that prone to Stress.

Hybrid segmentation of internet banking users

PurposeThe objective of this paper is to present and discuss the development of a transaction‐based model for segmenting users of internet banking. It aims to employ a random sample of clients of a large Canadian bank in generating the hybrid segments.Design/methodology/approachThe basic transactional profile of the bank's clients was merged with Mosaic's financial segments contained in the Generation5 database. A random sample of 3 percent of a large Canadian chartered bank's clients was drawn from its transaction database. The transaction database employed contains clients from Quebec and the Maritime provinces. The sampling frame consisted of close to one million clients. Two‐step cluster analysis was employed to generate the transaction segment and later merged with the Mosaic financial segment to produce hybrid segments.FindingsTwo‐step cluster analysis identified four generic transaction segments which, when cross‐tabulated with the Mosaic financial segments, produced highly stable and interpretable segments. These hybrid segments are clearly superior to conventional life style or psychographic segments produced by classical segmentation methodologies.Practical implicationsThe results of this study clearly demonstrate the functional and analytical superiority of hybrid customer segments. Hybrid segmentation, by cross‐tabulating transaction and Mosaic's financial segments, provides banks and financial institutions with superior strategic insights in customer understanding, customer segmentation, customer communication, customer prospecting and targeting.Originality/valueThis paper is the first to present, explain and to demonstrate the nature and the operational procedure to develop hybrid customer/client segments. More importantly, it is the first that goes beyond conventional approaches to segmenting banks' clients who engage in internet banking by integrating clients' transaction profiles and Mosaic financial segments. The resulting hybrid segments are radically different than the conventional, one‐dimensional segments produced by conventional cluster‐based segmentation.

HIV-patients discrimination according to phenotype and functional assay of T-cells subsets

Results Two step Cluster Analysis followed by a discriminant function analysis of the lymphocyte activation assay from the first blood sample, allowed the separation of HIV-1 patients in two groups: Cluster1 (67%) and Cluster2 (33%). Clusters definition relied on the level of three T-cells subsets: a) stimulated CD4CD69 CD25, b) unstimulated CD4CD69CD25 and c) unstimulated CD4CD25. PHA stimulated CD4 CD69 CD25 subset level alone allowed to classify correctly patients with 92% sensitivity and 87% specificity. Cluster-2 expressed more CD69 and HLA-DR activation markers on CD4 and CD8 lymphocytes, less CD8 CD28 and responded less to mitogen even if viral load undetectable. Cluster-2 presented poorly clinical profile in terms of previous AIDS events, current CD4 count, viral load, length of treatment. Over the time most patients (64%) were keeping their cluster category.

Whole life sustainability in the design of tourist resorts

PurposeThe purpose of this paper is to argue that it is possible to identify a number of physical design properties of plots that are associated with coastal alteration and to develop a predictive model or metric to evaluate the potential alteration caused by that resort. Investors may have economic plot size concerns not considered by planners, tourists may prefer close proximity to deep waters, and urban planners may have failed to realize the importance of certain plot ratios or minimum dimensions etc.Design/methodology/approachA combination of geographical information system and statistical tools is used to identify and model these properties. The research uses data from three different resorts in two countries along the Red Sea coast (Hurghada and Safaga in Egypt and Jeddah in Saudi Arabia). Landsat TM7 images are obtained for each location and analysed to detect regions with the largest alteration along the coast using the post‐classification change technique. Physical properties of each village are generated using ArcGIS 9.2 both before and after alteration. Finally, a predictive decision‐support tool for urban planners and environmental auditors is formulated using stepwise discriminant analysis.FindingsTwo‐step cluster analysis and discriminate analysis show that alteration takes place based on factors that include original width and original distance to deep water in three distinct groups. The predictive model that is developed classifies any given resort into one of three clusters, by virtue of their physical design characteristics. This gives a rough prediction of the likely alteration that might take place under a given condition. Subdivision plans that ignore the power of these two factors literally “invite” owners to violate shorelines and encroach on the sea. Planners are urged to join hands with policy makers and developers to formulate guidelines to plot subdivision and environmentally acceptable design solutions to facilitate access to deep water.Originality/valueTill now, there has been no metric to relate a plot's geometric properties with the encroachment it makes. If such an association could be made, a developer's resort plans may be evaluated to predict the degree by which an encroachment is likely. The model can be used to identify resorts whose properties present a risk to shorelines, dredge or fill to maximize usable land. Two benefits may be thought of for this approach. First, the model helps urban planners develop a sustainable coastal area by bearing in mind the needs of tourist‐developers as well as the properties of their adjacent coast. Second, the factors can be used to rationalize building guidelines and land use regulations to minimize such risk.

Understanding the Downshifting Phenomenon: A Case of South East Queensland, Australia

A quality of life survey of a sample of households across the Brisbane‐South East Queensland region has identified about 28 percent of people as ‘downshifters.‘ They are defined as people who voluntarily make a long‐term change in their lifestyle – other than planned retirement – which reduces income and adjusts their lifestyle conditions. A typology of downshifters is developed on the basis of their motives for downshifting and their socio‐economic and demographic characteristics using a Two Step Cluster Analysis. Results indicate that the social and economic circumstances and the reasons and methods of downshifting tend to vary substantially across the clusters.

요인분석과 군집분석을 통한 세분화 및 전략방향 제시: 특수법인 사례를 중심으로

세분화는 크게 목적변수의 유무에 따라 분석방법이 달라지게 된다. 본 논문은 특수법인의 경영지표를 바탕으로 목적변수가 존재하지 않을 경우의 세분화를 통해 전략방향을 도출하는 사례 연구를 제안하고자 한다. 군집분석을 통한 세분화의 경우, 많은 변수를 사용하여 분류를 하게 되면 군집별 특성화가 어렵게 된다. 따라서 군집의 특성을 잘 반영할 수 있는 대표적 요인변수를 요인분석을 통해 추출하고 이 대표요인을 바탕으로 2단계 군집분석을 통한 세분화를 고려하였다. 이를 통해 총 6개의 세분화 군집을 도출하고 각 군집 별 강점요인을 강화하고 약점요인을 보완하는 방향으로 전략방향을 설정하여 제안하고자 한다. Corporations adopt a segmentation depends on the existence of target variables, in general. In this paper, for the case of no target variables, a strategy through segmentation is proposed for corporations having a special status based on the management index. In case of segmentation using cluster analysis, however, if one classify according to many variables then he will be in face of difficulties in characterizing. Therefore, after extracting representative factors by factor analysis, a segmentation method through 2 step cluster analysis is employed on the basis of these representative factors. As a result, six segmentation groups are found and the resulting strategy is proposed which strengthens prominent factors and makes up defective factors for each group.

Quantitative Evaluation of Well characteristics by Step Drawdown Test and Cluster Analysis of Discharge and Water Level in Shallow Wells

Quantitative Evaluation of Well characteristics by Step Drawdown Test and Cluster Analysis of Discharge and Water Level in Shallow Wells

An approach to the construction of an immunoarray for differentiating and quantitating cross reacting analytes

We demonstrate the construction of an immunoarray to categorize, identify and quantitate different triazine herbicides or their environmental metabolites in water. In a first step cluster analysis is used to examine the performance of a small subset of antibodies out of a larger library, and hence to select a small array that is capable of categorizing the triazine herbicides into different groups. At the 1 ppb level it is possible to categorize an analyte as a chloro-s-triazine, a hydroxy metabolite, or as a methoxy/methylthio substituted triazine when using only 2–4 antibodies. At higher concentrations even the identity of the triazine can be determined with the same number of antibodies. A selected combination of antibodies is then used to identify and quantify unknowns by comparing their immunoassay responses to an array of calibration curves and using the maximum likelihood criterion.