Statistical tests for comparing two samples

2022 ◽  
pp. 156-178
Author(s):  
Vladimir Hedrih ◽  
Andjelka Hedrih
Keyword(s):  
Statistical Tests ◽  
Two Samples

scholarly journals Antecedents and outcomes of brand love: utilitarian and symbolic brand qualities

2018 ◽  
Vol 52 (3/4) ◽  
pp. 656-682 ◽  
Author(s):  
Cristela Maia Bairrada ◽  
Filipe Coelho ◽  
Arnaldo Coelho
Keyword(s):  
Structural Equation ◽  
Statistical Tests ◽  
Information Processing Model ◽  
Cross Sectional ◽  
Brand Love ◽  
Content Type ◽  
Two Samples ◽  
Premium Price ◽  
The Relationship ◽  
Practical Implications

Purpose Brand love is associated with consumer behaviours that are key for organisational performance. However, research on the antecedents of brand love is sparse. The current research draws on the information processing model as well as on the experiential approaches to consumer behaviour to develop a model comprising a novel set of antecedents. Design/methodology/approach To test the research hypotheses, we resort to two samples, which implied the collection of usable 1,018 questionnaires. For hypotheses testing, we resort to structural equation modelling. Findings Both functional constructs as well as more symbolic/emotional ones are positively associated with brand love. In addition, constructs with a more functional nature tend to have an indirect effect on brand love, whereas constructs with a higher level of abstraction tend to mediate the effects of more specific brand qualities. Finally, brand love is related with important outcomes, including loyalty, word of mouth and willingness to pay a premium price. Research/limitations implications This research has a cross-sectional nature. Moreover, we rely on a single informant, but the procedural remedies as well as the statistical tests we conducted suggest that common method variance is not a concern. Practical implications The findings suggest that managers should emphasise both functional as well as emotional/symbolic aspects to strengthen the links between brands and consumers, which will be beneficial for both sides. Originality/value This study is the first to investigate the relationship between a number of symbolic and functional brand aspects and the development of brand love feelings.

The Evidential Value of Multiple Continuous Measurements — A Simplified Approach to Data Analysis

1980 ◽  
Vol 20 (3) ◽  
pp. 154-162 ◽  
Author(s):  
K.W. Smalldon ◽  
C. Brown
Keyword(s):  
Data Analysis ◽  
Statistical Tests ◽  
Simple Approach ◽  
Optimum Performance ◽  
Simplified Approach ◽  
Evidential Value ◽  
Discriminating Power ◽  
Test Variable ◽  
Two Samples ◽  
Representative Samples

A conceptually simple approach is described for the interpretation of all types of multidimensional continuously variable data. Items which reflect the problem being investigated are collected and two independent representative samples are taken from each item for analysis. A single calculated test variable is used, in conjunction with a computer and the analyst's intuitive judgment, to interpret the data. Although optimum performance cannot be guaranteed the method is flexible and allows assessment of two samples for similarity, the calculation of discriminating power for any set of measurements and estimation of the discrimination achieved from the analysis of a particular item. Since formal statistical tests and tables are not used at any stage, analysts are competent to investigate their own data after only a little practice. The system is equally suited to research and operational requirements. Some applications and potential developments of the system are discussed.

Statistical tests for the comparison of two samples: The general alternative

2016 ◽  
Vol 46 (2) ◽  
pp. 903-909 ◽  
Author(s):  
Markus Neuhäuser ◽  
Anke Welz ◽  
Graeme D. Ruxton
Keyword(s):  
Statistical Tests ◽  
Two Samples ◽  
General Alternative

scholarly journals Enzyme-Linked Immunosorbent Assay Kit for Beta-Lactoglobulin Determination: Interlaboratory Study

2009 ◽  
Vol 92 (5) ◽  
pp. 1519-1525 ◽  
Author(s):  
František Štumr ◽  
Dana Gabrovská ◽  
Jana Rysová ◽  
Petr Hanák ◽  
Jan Plicka ◽  
...  
Keyword(s):  
Wheat Flour ◽  
Statistical Tests ◽  
Milk Proteins ◽  
Food Samples ◽  
Chicken Meat ◽  
Interlaboratory Study ◽  
Enzyme Linked Immunosorbent Assay ◽  
Study Results ◽  
Two Samples ◽  
Beta Lactoglobulin

Abstract An interlaboratory study was performed in six laboratories to prove the validation of the ELISA method developed for quantitative determination of beta-lactoglobulin (BLG) in foods. The ELISA kit used for this study is based on rabbit polyclonal antibody. In-house validation of the kit did not produce false-positive results or cross-reactivity in a broad range of food matrixes containing no milk proteins. All participants obtained the BLG kit with a standard operational procedure, the list of the samples, samples, and a protocol for recording test results. The study included 14 food samples (extruded breakfast cereals, bread, two soy desserts, butter, chicken ham, chicken meat, wheat flour, long grain rice, jelly, two whey drinks, crackers, and bitter chocolate) and six spiked samples (two rice, two wheat flour, and two chicken meat). Nine samples of food matrixes containing no milk proteins showed BLG content lower than the first standard (0.15 mg/kg). Two samples of food matrixes with no milk proteins revealed BLG content higher than standard 3 (1.5 mg/100 g) and standard 4 (5.0 mg/100 g). Three food samples containing milk were tested as positive, and all spiked samples were evaluated as positive. The statistical tests (Cochran, Dixon, and Mandel) and analysis of variance were used to evaluate the interlaboratory study results. Repeatability and reproducibility limits, as well as LOQ (0.22 mg BLG/kg) and LOD (0.07 mg BLG/kg), for the kit were calculated.

scholarly journals Are Islamic Banks Really More Solvent Than Conventional Banks in a Financial Stable Period?

2019 ◽  
Vol 11 (2) ◽  
pp. 15 ◽  
Author(s):  
Achraf HADDAD ◽  
Anis EL AMMARI ◽  
Abdelfettah BOURI
Keyword(s):  
Statistical Tests ◽  
Stock Exchanges ◽  
Islamic Banks ◽  
Stable Period ◽  
Banking Market ◽  
Two Samples ◽  
Operative Evaluation ◽  
Conventional Banks ◽  
The Comparative Study ◽  
Conventional And Islamic Banks

The knowledge value produced by this research was established in particular by the methodological challenges of the comparative study. Based on a process of bibliographic research, available conditional observation and necessary statistical tests, we innovated an equiprobable comparison between the solvencies of conventional and Islamic banks over the period (2010-2018). Our study is not only a matter of dealing generically with the financial solvency question of conventional and Islamic banks, but also, we analysed the inherent implications and difficulties that may alter the results and influence the establishment of an operative evaluation of financial institutions. Two samples were taken from two reference populations. The core populations are all existing conventional and Islamic banks in the selected countries. The choice of banks is limited to countries whose banking systems incorporate both Islamic and conventional banks regardless of the proportion of each system in the country's banking market. Subsequently, each list bank was reduced on the basis of qualitative and quantitative filtering criteria. Therefore, each conventional bank has its closest Islamic equivalence in terms of capital and size taken from the same country. This restriction reduced the sample size to 63 banks each. The selected banks are all large and listed in different stock exchanges around the world. In conclusion, we found that conventional banks are more solvent than Islamic banks during a financial stable period.

scholarly journals Elemental Characterization of Ciders and Other Low-Percentage Alcoholic Beverages Available on the Polish Market

Molecules ◽  
2021 ◽  
Vol 26 (8) ◽  
pp. 2186
Author(s):  
Magdalena Gajek ◽  
Aleksandra Pawlaczyk ◽  
Piotr Wysocki ◽  
Malgorzata I. Szynkowska-Jozwik
Keyword(s):  
Raw Materials ◽  
Statistical Tests ◽  
Analytical Data ◽  
Alcoholic Beverages ◽  
Data Set ◽  
Apple Variety ◽  
Three Samples ◽  
Two Samples ◽  
Elemental Fingerprint ◽  
The Impact

Seventy-three samples of alcoholic beverages and juices that were purchased on the Polish market and home-made were analyzed for their elemental profiles. The levels of 23 metals were determined by ICP-MS (Ag, Ba, Bi, Cd, Co, Cr, Li, Mn, Ni, Pb, Sr and Tl), ICP-OES (Al, B, Ca, Cu, Fe, K, Mg, Na, Ti and Zn) and CVAAS (Hg) techniques in twenty-five samples of ciders widely available on the Polish market; six samples of home-made ciders; two samples of juices used in the production of these ciders; and forty samples of low-percentage, flavored alcoholic beverages based on beer. The gathered analytical data confirmed that the final elemental fingerprint of a product is affected by the elemental fingerprint of the ingredients used (apple variety) as well as the technology and equipment used by the producer, and in the case of commercial ciders, also the impact of type of the packaging used was proven. These factors are specific to each producer and the influence of the mentioned above parameters was revealed as a result of the performed analysis. Additionally, the inclusion of the home-made ciders in the data set helped us to understand the potential origin of some elements, from the raw materials to the final products. The applied statistical tests revealed (Kruskal–Wallis and ANOVA) the existence of statistically significant differences in the concentration of the following metals: Ag, Al, B, Bi, Co, Cr, Cu, Fe, K, Li, Mg, Na, Ni, Ti and Zn in terms of the type of cider origin (commercial and home-made). In turn, for different packaging (can or bottle) within one brand of commercial cider, the existence of statistically significant differences for Cu, Mn and Na was proved. The concentrations of all determined elements in the commercial cider from the Polish market and home-made cider samples can be considered as nontoxic, because the measured levels of elements indicated in the regulations were lower than the allowable limits. Moreover, the obtained results can be treated as preliminary for the potential authentication of products in order to distinguish the home-made (fake) from the authentic products, especially for premium-class alcoholic beverages.

scholarly journals Application of the T-Test in Health Insurance Cost Analysis: Large Data Sets

ECONOMICS ◽  
2019 ◽  
Vol 7 (2) ◽  
pp. 157-167
Author(s):  
Bojan Kresojević ◽  
Milica Gajić
Keyword(s):  
Health Insurance ◽  
Statistical Tests ◽  
Bootstrap Method ◽  
T Test ◽  
Risk Level ◽  
P Value ◽  
Whitney Test ◽  
Mann Whitney Test ◽  
Insurance Cost ◽  
Two Samples

Abstract In this paper will be analyzed the application of the t-test against the nonparametric Mann - Whitney test in the analysis of health insurance benefit costs in the Republic of Srpska on large samples. This research aims to examine which method produces better results when testing statistical hypotheses. The adequacy of the statistical tests will be tested on primary health insurance cost data for 1,044,690 insureds in 2017. For two samples of size 4,000, the sampling distribution of the difference in two means has a skewness coefficient of 0.05 and a kurtosis coefficient of 3.09. Jarque - Bera test does not reject the hypothesis of normality of distribution with a p-value of 0.135. On the other hand, in the Mann - Whitney test, the real risk of the first species, when there is a difference in skewness between the samples, may be less than 0.001 compared to the nominal risk level of 0.05. Based on the results obtained, it is suggested to use the t-test instead of the Mann - Whitney test if the sample is large enough, which should be verified by the bootstrap method.

scholarly journals IMPROVEMENT EVALUATION ON CERAMIC ROOF EXTRACTION USING WORLDVIEW-2 IMAGERY AND GEOGRAPHIC DATA MINING APPROACH

2016 ◽  
Vol XLI-B7 ◽  
pp. 883-889
Author(s):  
V. S. Brum-Bastos ◽  
B. M. G. Ribeiro ◽  
C. M. D. Pinho ◽  
T. S. Korting ◽  
L. M. G. Fonseca
Keyword(s):  
Data Mining ◽  
High Resolution ◽  
Near Infrared ◽  
Statistical Tests ◽  
Classification Tree ◽  
Semantic Networks ◽  
Urban Environments ◽  
Urban Land Use ◽  
Two Samples ◽  
High Resolution Images

Advances in geotechnologies and in remote sensing have improved analysis of urban environments. The new sensors are increasingly suited to urban studies, due to the enhancement in spatial, spectral and radiometric resolutions. Urban environments present high heterogeneity, which cannot be tackled using pixel–based approaches on high resolution images. Geographic Object–Based Image Analysis (GEOBIA) has been consolidated as a methodology for urban land use and cover monitoring; however, classification of high resolution images is still troublesome. This study aims to assess the improvement on ceramic roof classification using WorldView-2 images due to the increase of 4 new bands besides the standard “Blue-Green-Red-Near Infrared” bands. Our methodology combines GEOBIA, C4.5 classification tree algorithm, Monte Carlo simulation and statistical tests for classification accuracy. Two samples groups were considered: 1) eight multispectral and panchromatic bands, and 2) four multispectral and panchromatic bands, representing previous high-resolution sensors. The C4.5 algorithm generates a decision tree that can be used for classification; smaller decision trees are closer to the semantic networks produced by experts on GEOBIA, while bigger trees, are not straightforward to implement manually, but are more accurate. The choice for a big or small tree relies on the user’s skills to implement it. This study aims to determine for what kind of user the addition of the 4 new bands might be beneficial: 1) the common user (smaller trees) or 2) a more skilled user with coding and/or data mining abilities (bigger trees). In overall the classification was improved by the addition of the four new bands for both types of users.

scholarly journals IMPROVEMENT EVALUATION ON CERAMIC ROOF EXTRACTION USING WORLDVIEW-2 IMAGERY AND GEOGRAPHIC DATA MINING APPROACH

2016 ◽  
Vol XLI-B7 ◽  
pp. 883-889 ◽  
Author(s):  
V. S. Brum-Bastos ◽  
B. M. G. Ribeiro ◽  
C. M. D. Pinho ◽  
T. S. Korting ◽  
L. M. G. Fonseca
Keyword(s):  
Data Mining ◽  
High Resolution ◽  
Near Infrared ◽  
Statistical Tests ◽  
Classification Tree ◽  
Semantic Networks ◽  
Urban Environments ◽  
Urban Land Use ◽  
Two Samples ◽  
High Resolution Images

Advances in geotechnologies and in remote sensing have improved analysis of urban environments. The new sensors are increasingly suited to urban studies, due to the enhancement in spatial, spectral and radiometric resolutions. Urban environments present high heterogeneity, which cannot be tackled using pixel–based approaches on high resolution images. Geographic Object–Based Image Analysis (GEOBIA) has been consolidated as a methodology for urban land use and cover monitoring; however, classification of high resolution images is still troublesome. This study aims to assess the improvement on ceramic roof classification using WorldView-2 images due to the increase of 4 new bands besides the standard “Blue-Green-Red-Near Infrared” bands. Our methodology combines GEOBIA, C4.5 classification tree algorithm, Monte Carlo simulation and statistical tests for classification accuracy. Two samples groups were considered: 1) eight multispectral and panchromatic bands, and 2) four multispectral and panchromatic bands, representing previous high-resolution sensors. The C4.5 algorithm generates a decision tree that can be used for classification; smaller decision trees are closer to the semantic networks produced by experts on GEOBIA, while bigger trees, are not straightforward to implement manually, but are more accurate. The choice for a big or small tree relies on the user’s skills to implement it. This study aims to determine for what kind of user the addition of the 4 new bands might be beneficial: 1) the common user (smaller trees) or 2) a more skilled user with coding and/or data mining abilities (bigger trees). In overall the classification was improved by the addition of the four new bands for both types of users.

A quantum leap in the reproducibility, precision, and sensitivity of gene expression profile analysis even when sample size is extremely small

2015 ◽  
Vol 13 (04) ◽  
pp. 1550018 ◽  
Author(s):  
Kevin Lim ◽  
Zhenhua Li ◽  
Kwok Pui Choi ◽  
Limsoon Wong
Keyword(s):  
Gene Expression ◽  
Sample Size ◽  
Small Sample Size ◽  
Statistical Tests ◽  
Transcript Level ◽  
Small Sample ◽  
Sample Sizes ◽  
Quantum Leap ◽  
Two Samples ◽  
Small Sample Sizes

Transcript-level quantification is often measured across two groups of patients to aid the discovery of biomarkers and detection of biological mechanisms involving these biomarkers. Statistical tests lack power and false discovery rate is high when sample size is small. Yet, many experiments have very few samples (≤ 5). This creates the impetus for a method to discover biomarkers and mechanisms under very small sample sizes. We present a powerful method, ESSNet, that is able to identify subnetworks consistently across independent datasets of the same disease phenotypes even under very small sample sizes. The key idea of ESSNet is to fragment large pathways into smaller subnetworks and compute a statistic that discriminates the subnetworks in two phenotypes. We do not greedily select genes to be included based on differential expression but rely on gene-expression-level ranking within a phenotype, which is shown to be stable even under extremely small sample sizes. We test our subnetworks on null distributions obtained by array rotation; this preserves the gene–gene correlation structure and is suitable for datasets with small sample size allowing us to consistently predict relevant subnetworks even when sample size is small. For most other methods, this consistency drops to less than 10% when we test them on datasets with only two samples from each phenotype, whereas ESSNet is able to achieve an average consistency of 58% (72% when we consider genes within the subnetworks) and continues to be superior when sample size is large. We further show that the subnetworks identified by ESSNet are highly correlated to many references in the biological literature. ESSNet and supplementary material are available at: http://compbio.ddns.comp.nus.edu.sg:8080/essnet .

Sign in / Sign up

Export Citation Format

Share Document