Experimental Statistical Analysis Research Articles

On the Statistical Analysis of Experiments With Manipulation Checks

Psychologists are often interested in the effect of an internal state, such as ego depletion, that cannot be directly assigned in an experiment. Instead, they assign participants to a manipulation intended to produce this state and use manipulation checks to assess the manipulation’s effectiveness. In this article, I discuss statistical analyses for experiments in which researchers are primarily interested in the average treatment effect (ATE) of the target internal state rather than that of the manipulation. Often, researchers estimate the association of the manipulation itself with the dependent variable, but this intention-to-treat (ITT) estimator is typically biased for the ATE of the target state, and the bias could be either toward the null (conservative) or away from the null. I discuss the fairly stringent assumptions under which this estimator is conservative. Given this, I argue against the status-quo practice of interpreting the ITT estimate as the effect of the target state without any explicit discussion of whether these assumptions hold. Under a somewhat weaker version of the same assumptions, one can alternatively use instrumental-variables (IVs) analysis to directly estimate the effect of the target state. IVs analysis complements ITT analysis by directly addressing the central question of interest. As a running example, I consider a multisite replication study on the ego-depletion effect, in which the manipulation’s partial effectiveness led to criticism and several reanalyses that arrived at varying conclusions. I use IVs analysis to directly account for the manipulation’s partial effectiveness; this corroborated the replication authors’ reported null results.

A Multi-Cycle Echo Energy Concentration Method for High-Mobility Targets Enveloped by Time-Varying Plasma Sheath

When a target moves at hypersonic speed, the aerodynamic thermal effect will cause air molecules to form a plasma sheath that envelopes the outer surface of the target, which consists of a large number of charged particles. The plasma sheath imposes a complicated modulation effect on the radar echo signal in terms of amplitude, phase, and frequency. When the plasma sheath is time-varying, the inter-pulse coherence of the multi-cycle echo signals is severely disrupted, resulting in the failure of coherent accumulation. To address the problem of abnormal inter-pulse energy accumulation in targets covered with time-varying plasma sheaths, we analyzed the dynamic modulation effects of time-varying plasma sheaths on echo signals and constructed a radar echo model enveloped with time-varying plasma sheaths. Based on this, we propose a method for inter-pulse energy concentration of multi-cycle echo signals based on range-frequency inversion, second-order Wigner–Ville distribution (WVD), and slow-time symmetric auto-correlation. The proposed method is capable of realizing energy concentration for targets enveloped with time-varying plasma sheaths and can accurately estimate the motion parameters of the target. The effectiveness of our proposed method has been verified via simulation analysis of multi-cycle echo signals from targets enveloped with time-varying plasma sheaths, and the reliability of the method has been further validated through statistical experimental analysis.

TOWARDS EYE-TRACKING-BASED TECHNOLOGY ON SIGHT INTERPRETATION PERFORMANCE IMPROVEMENT

This study used the recording of eye-tracking-based target domain (TD) fixation as the primary approach to explore the correlation between occupying fixation and sight interpretation (SI) performance. This paper records the gaze plot and gaze duration during the sight interpretation and analyzes the correlation between them and the interpretation performance. First, we designed the nine-point track calibration for a noninvasive study. Second, we carried out pre-experiments to find out the best experimental conditions. Finally, after eye rest, we performed the formal test. Extensive experiments were implemented to verify the factors that affect the SI performance, including the number of TD occupying fixations, the time-cost of TD occupying fixation, and the concentration of TD occupying fixation. Statistical analysis of experiments concluded that the psychological dictionary, Long-term Memory (LTM) information, and bilingual conversion skills are the main factors affecting the number and time of eye-tracking TD occupying fixation spots.

PREFACE

PREFACE

The generation of the experimental design

Texts and teaching of Experimental Statistics emphasize the statistical analysis of experiments and make only references to the conceptual foundation of experimental research. Basic concepts are vaguely and incompletely defined, and experimental designs are presented as a set of recipes from which one must be chosen for each particular experiment. Consequently, inferences derived from the experiment are often biased. A conceptual basis for the experimental research is considered and a rational procedure for generating the experimental design is suggested. The generation of the experimental design is based on the separate definitions of the structures of the experimental factors and the unit factors, and the association of these two structures by randomization and presence of the experimental factors in the sample. This approach leads to the clear identification of the confounding of the effects from these two structures and of the experimental errors that affect the effects of the experimental factors.

Validation and usability of AIDMED - telemedical system for cardiological and pulmonary diseases

Aim: The aim of this paper is to present the results of the validation of AIDMED as a telemedical system, i.e. its capability in faithful registration of biomedical signals, its acquisition in a telemedical scenario and its representation in online application. Usability of sucha tool for a dedicated population was also assessed.Methods: We describe and discuss functionalities provided by AIDMED. We perform a series of experiments where we measure biological signals with AIDMED and with a reference device. We provide statistical analysis of experiments. We also compare the functionality of AIDMED with other similar solutions. We discuss the usability of AIDMED in tele observation of COVID-19 patients.Results: We show diagnostic equivalence of AIDMED device and reference devices.Moreover, we indicate advantages of AIDMED system (as task management and patient’s feedback via mobile app) for at home telemonitoring in comparison to standard of care.Conclusions: AIDMED system provides an integrated platform which enables observation of COVID-19, cardiological and pulmonary patients and many more. Thus, an opportunity for both better quality of care and better subjective patient satisfaction with use of AIDMED has got a solid foundation.

Constructing non-orthogonal split-split-plot designs using some resolvable block designs

Summary We consider a new method of constructing non-orthogonal (incomplete) split-split-plot designs (SSPDs) for three (A, B, C) factor experiments. The final design is generated by some resolvable incomplete block design (for the factor A) and by square lattice designs for factors B and C using a modified Kronecker product of those designs (incidence matrices). Statistical properties of the constructed designs are investigated under a randomized-derived linear model. This model is strictly connected with a four-step randomization of units (blocks, whole plots, subplots, sub-subplots inside each block). The final SSPD has orthogonal block structure (OBS) and satisfies the general balance (GB) property. The statistical analysis of experiments performed in the SSPD is based on the analysis of variance often used for multistratum experiments. We characterize the SSPD with respect to the stratum efficiency factors for the basic estimable treatment contrasts. The structures of the vectors defining treatment contrasts are also given.

Development, implementation and optimization of a mobile 3D printing platform

One of the biggest challenges to 3D printing is that typical desktop printers are stationary with a limited workspace. A mobile 3D-printing platform, which has omnidirectional wheels that allow for unrestricted movement along x- and y-axes, can alleviate that restriction. The research team in this project performed a series of preliminary material tests with such a fully constructed platform. The system was tested with materials from three different industries that could benefit from mobile additive manufacturing technology. A cement paste was tested for the construction industry, frosting for the food industry, and clay paste for the fine arts industry. Next, a statistical experimental analysis was performed to determine the optimum printing parameters to obtain geometrical accuracy of the object being printed. The independent variables chosen for the material and the printing platform included the material type, percent concentration of dry material to wetting agent, layer height, layer width, and printing speed. Multiple samples were printed for each combination of independent variables. The dependent variables, maximum taper angle and x-y-z measurements, were then found from the printed samples. Principal Component Analysis was performed on the taper angle and x-y-z measurements to create a single index which represented the error in print quality. A slightly modified 2 k factorial design was then used to determine which printing parameters and material type significantly affected the error index. Finally, response surface methodology along with the method of steepest ascent was used to identify the optimum printing parameters.

Communicating statistical conclusions of experiments to scientists

AbstractThe manuscript introduces a framework for presenting the results of the statistical analysis of experiments with multiple responses and multiple factors. We propose a utilisation of factors scaling to enable a transformation that combines main effects, quadratic effects and interactions into a meaningful summary that allows the scientist/experimenter to immediately recognise the most influential factors for a given response. The framework does not replace the thorough evaluation of the results but provides a clear high‐level summary of the relative importance of findings. The visualisation of such factor importance, using intensity heatmaps, allows the immediate understanding of the results across multiple responses that efficiently guides a following detailed analysis of certain responses and factors and contributes in designing subsequent experiments. The methodology is applied to a real industrial experiment and to a simulated data set with a larger number of responses and factors.

Experiment: Conceptual Basis

Texts and teaching of Experimental Statistics emphasize the statistical analysis of experiments and superficially consider the conceptual basis of experimental research. Definitions of basic concepts are imprecise, incoherent and ambiguous. This is the case, for example, with the concepts of experimental material, experimental factor, experimental unit and experimental error. In particular, the lack of distinction between experimental factor and unit factor and between the two classes of experimental factor: treatment factor and intrinsic factor leads to flaws in the plan and analysis of experiments that originate biased inferences. This approach gives rise to ignorance and misunderstanding of these concepts, underestimation of the importance of the planning of the experiment and, consequently, the inefficiency of many research. This paper reviews and reformulates important concepts with the purpose of contributing a rational basis for experimental research and, in particular, for Experimental Statistics. These revised concepts provides a basis for the formulation of an experiment structure that provides appropriate inferences for the achievement of the objectives of the experiment.

Theoretical and Experimental Analysis of Cryptographic Hash Functions

The paper presents a theoretical introduction to the cryptographic hash function theory and a statistical experimental analysis of selected hash functions. The definition of hash functions, differences between them, their strengths and weaknesses are explained as well. Different hash function types, classes and parameters are described. The features of hash functions are analyzed by performing statistical analysis. Experimental analysis is performed for three certified hash functions: SHA1-160, SHA2-512 and SHA3-512. Such an analysis helps understand the behavior of cryptographic hash functions and may be very helpful for comparing the security level of the hashing method selected. The tests may serve as a basis for examination of each newly proposed hash function. Additionally, the analysis may be harness as a method for comparing future proposals with the existing functions.

A multi-objective evolutionary algorithm based on Pareto prediction for automatic test task scheduling problems

The test task scheduling problem (TTSP) is a combinatorial optimization problem still under investigation. A multi-objective evolutionary algorithm based on Pareto prediction (PP-MOEA) is proposed fully considering the characteristics of TTSP. In a multi-objective TTSP, multiple solutions in the decision space correspond to a point in the objective space and the number of true Pareto front in the objective space is relatively small. Moreover, there are less solutions distributing in either high or low objective values while most solutions distribute in average. Based on these characteristics, a pure elitism strategy and a novel scale-down non-dominated sorting (SDNS) method are combined to improve the efficiency and convergence of the algorithm. Additionally, an extended encoding range approach and a Pareto prediction strategy are proposed to help exploring new solutions and increase the diversity of population. The Pareto prediction strategy is implemented by following the historical evolutionary information of the current Pareto set and then predicting the next generation of several individuals. The makespan (maximal test completion time) and the mean workload of the instruments are considered in this research. Based on the general framework of MOEAs, the proposed algorithm is more effective and efficient in tackling multi-objective TTSPs. PP-MOEA has the better performance of convergence and diversity compared with other algorithms based on the statistical analysis of experiments results.

Combination of gravity concentration variables to increase the productivity of the Brucutu mineral processing plant

The jigging process is one of the oldest ways used to concentrate ores in the world, suitable for small and medium-sized employability, and is composed of a pre-concentration step as a way to reduce costs. Within this perspective, a test campaign for laboratorial mineral concentration was set up, in order to determine the main variables (and their interactions) that may have an influence on the concentration of iron ore from the Quadrilátero Ferrífero by the jigging process. A factorial planning was prepared so that an accurate statistical analysis of experiments could be performed. A combination of the three variables were studied (feed granulometry, background layer, and bed mass) promoted an iron recovery equal to 73.14%, a distribution of silica of 89.05% on the tailing and a selectivity index of 3.80. Thus, the methodology enabled a useful statistical evaluation and can contribute to decision-making regarding the parameters of interest in the jigging process.

Using InVivoStat to perform the statistical analysis of experiments.

The need to improve reproducibility and reliability of animal experiments has led some journals to increase the stringency of the criteria that must be satisfied before manuscripts can be considered suitable for publication. In this article we give advice on experimental design, including minimum group sizes, calculating statistical power and avoiding pseudo-replication, which can improve reproducibility. We also give advice on normalisation, transformations, the gateway analysis of variance strategy and the use of p-values and confidence intervals. Applying all these statistical procedures correctly will strengthen the validity of the conclusions. We discuss how InVivoStat, a free-to-use statistical software package, which was designed for life scientists, especially animal researchers, can be used to help with these principles.

Experimental Determination of the Influence of an Air Entraining Additive on the Resistance of Concrete to Chemical Defrosting Agents

The paper describes the results of an experiment performed as part of the GAČR 13-18870S project dealing with the durability characteristics of the surface layer of hardened concrete. The paper aims to evaluate the experimental determination of the influence of an air entraining additive on the resistance of cement concrete to water and defrosting chemicals. For the purposes of the experiment 4 mixtures were prepared which differed in the amount of air entraining additive and the amount of cement. The test results were evaluated using statistical analysis of experiment.

Evaluation of the Antifungal Activity of Sulfur and Chitosan Nanocomposites with Active Ingredients of Ruta graveolens, Thymus vulgaris and Eucalyptus melliodora on the Growth of Botrytis fabae and Fusarium oxysporum

The purpose of this research was to determine the inhibitor effects of nanocomposites over two pathogenic fungi: Botrytis fabae Sard and Fusarium oxysporum isolated from bean crops. These nanocomposites were prepared with elemental sulfur nanoparticles coated with essential oils (four levels, three oils) and further stabilized with chitosan. The research was conducted in the microbiology laboratory of CENCINAT. We evaluated 12 nanocomposites at room temperature and PDA plates at pH 4 for 10 days and compared to controls with no nanocomposites additions. Data for statistical analysis were gathered starting from the seventh growth day because of the fungi efficient growth without treatment is between the fourth and fifth day, respectively. Statistical experimental analysis were carried out through a complete randomized block design using each Petri dish, with spots with 4 nacomposites in each spot. A total of 12 final different nanocomposites were used for all assayed plates giving five repetitions, four replicates and two fungi to carry out ANOVA analysis. Significant differences measures on fungal growth in each treatment, were compared according to Docimo Dunca's multiple range for p <0.05 using the statistical package InfoStat Professional. The independent analyzed variables were Fusarium oxysporum, Botrytis fabae Sard and different essential oils produced by Ruta graveolens, Eucalyptus melliodora, Thymus vulgaris and concentrations of each oil (20%, 40%, 60%. 80%). The results indicate that the nanocomposites application over the fungi in vitro could be effective according to the fungi species. F. oxysporum is the most inhibited by applying the nanocomposite with common rue or thyme oil 40% (optimum concentration), while 60% and 80% did not show any statistical difference. However, B. fabae inhibition was shown at these 60 and 80% concentrations. The nanocomposites with 20% eucalyptus displayed smaller inhibition halos against both fungi.

Analysis of an unreplicated 22 factorial experiment performed in a continuous process

This paper presents a tentative analysis method for unreplicated factorial designs where regular statistical experimental analysis cannot be used. The methodology is demonstrated through the analysis of an unreplicated two-level, two-factor factorial experiment performed in a continuous production process where the process was not in statistical control and where changes in the experimental design made conventional experimental analysis impossible. The first step of the analyses included screening of the sampled data. Principal component analysis and factor analysis were then used to create an overview of how the various responses and experimental factors were related. Carbon monoxide efficiency was selected as the most important parameter to be analysed further. Elastic net regression was used as a screening tool to remove non-significant factors, interaction, and covariates. Finally, the carbon monoxide efficiency variation was modelled using an intervention analysis. Two experimental factors were found to actively influence the response. The experiment that from other perspectives can be considered to be unanalysable, did thus reveal causal effects. The results imply that for processes where the process dynamics may be monitored, observations of the process dynamics may reduce the needs for repeated experimental runs, thus reducing the experimental costs.

The impact of particle size and initial solid loading on thermochemical pretreatment of wheat straw for improving sugar recovery

This work studies the effect of initial solid load (4-32 %; w/v, DS) and particle size (0.41-50 mm) on monosaccharide yield of wheat straw subjected to dilute H(2)SO(4) (0.75 %, v/v) pretreatment and enzymatic saccharification. Response surface methodology (RSM) based on a full factorial design (FFD) was used for the statistical analysis of pretreatment and enzymatic hydrolysis. The highest xylose yield obtained during pretreatment (ca. 86 %; of theoretical) was achieved at 4 % (w/v, DS) and 25 mm. The solid fraction obtained from the first set of experiments was subjected to enzymatic hydrolysis at constant enzyme dosage (17 FPU/g); statistical analysis revealed that glucose yield was favored with solids pretreated at low initial solid loads and small particle sizes. Dynamic experiments showed that glucose yield did not increase after 48 h of enzymatic hydrolysis. Once established pretreatment conditions, experiments were carried out with several initial solid loading (4-24 %; w/v, DS) and enzyme dosages (5-50 FPU/g). Two straw sizes (0.41 and 50 mm) were used for verification purposes. The highest glucose yield (ca. 55 %; of theoretical) was achieved at 4 % (w/v, DS), 0.41 mm and 50 FPU/g. Statistical analysis of experiments showed that at low enzyme dosage, particle size had a remarkable effect over glucose yield and initial solid load was the main factor for glucose yield.

SUPERCRITICAL FLUID EXTRACTION OF α - MANGOSTIN FROM MANGOSTEEN FRUIT RIND (GARCINIA MANGOSTANA LINN.)

The extraction of α-mangostin from mangosteen fruit rind Garcinia mangostana Linn cultivated in Vietnam using supercritical carbon dioxide with ethanol as a modifier was studied in this work. Experimental design was used to investigate the effect of operating pressure (20-30MPa), temperature (313-333 K), and solvent flow rate (10-20g/min) on the recovery and the selectivity of α-mangostin. Statistical analysis of experiments indicated that pressure and flow rate of solvent affected on the recovery of α-mangostin; with the selectivity of α-mangostin, pressure, solvent flow rate, temperature and interaction between these factors significantly affected. The central composite design showed that the polynomial regression models were in good agreement with the experimental results with the coefficients of determination from 0.9037 to 0.9666. Based on the proposed model, the optimal condition for extraction that was within the experimental range was found to be at lowest temperature 313 K, highest pressure 30 MPa and the nearly lowest flow rate of solvent 12.71 g/min. It was proved that the model equation developed can be used for predicting the extraction of α-mangostin from Garcinia mangostana Linn. The maximum recovery and selectivity of α-mangostin achieved at the optimal conditions was 22.83 mg/g material and 32.77%.

Preparation of alginate and chitosan nanoparticles using a new reverse micellar system

Alginate and chitosan nanoparticles were prepared using a new reverse micelle system, composed of cetyltrimethylammonium bromide (CTAB) as a surfactant, isooctane as a solvent, and 1-hexananol as a co-solvent. The obtained nanoparticles were characterized by FTIR, DLS and TEM techniques. The main objective of this study was to investigate the effects of polymer concentration, water content, and volumetric ratio of co-solvent to solvent on the physical and morphological properties of the prepared nanoparticles. To evaluate the results, the design of experimental was initially carried out and then the obtained data were statistically analyzed using the Qualitek-4® software. Results revealed that the size of the prepared alginate and chitosan nanoparticles varied in the range 220–490 and 210–1,050 nm, respectively. Furthermore, increasing either alginate or chitosan concentration increased the size of their nanoparticles. The results also showed that the size of nanoparticles was decreased with increasing the volumetric ratio of co-solvent/solvent. Finally, the size of alginate nanoparticles was increased by increasing the water content while it decreased the size of chitosan nanoparticles. Considering the statistical analysis of experiments, the polymer concentration is the major parameter affecting nanoparticles’ size. In contrast, water content has the smallest effect on the size of nanoparticles. However, the difference between the particle sizes of chitosan and alginate nanoparticles cab be attributed to the electrostatic repulsion between chitosan and CTAB.