Scott-Knott is an hierarchical clustering algorithm used in the ANOVA context, when researcher comparing treatment means, with a very important characteristic: it does not present any overlapping its grouping results. We wrote code, R, that performs this starting from vectors, data.frame, aov or aov.list objects. The results are presented letters representing groups as well graphical way using different colors to differentiate among distinct groups.<br /><br />

Regression Tree Method is not yet a mainstream method in Education, despite of being traditional approach Machine Learning. We advocate that this should become since, our point view, it the most suitable to analyse complex datasets, very common Education. This is, for example, case educational governmental large-scale databases, particular those where information: (1) have large quantity and types variables; (2) exhibit many categorical variables with categories; (3) non-linear relationships...

In this article, we present and argue our assertion that current routine psychological testing of individuals is not valid. To support assertion, review the concept ergodicity, Birkhoff’s theorem, Molenaar’s manifesto, which together contention direct transposition population estimations for producing inferences about individual We practice root cause why then provide an example a common application individual, explaining Finally, discuss how intraindividual (or within-person) approach...

Abstract The Learning Approaches Scale (EABAP) showed evidence of structural and external validity in assessing the deep surface approach elementary high school students. However, this is supported only by participants from a single school. present study evaluates generality EABAP verifying through multigroup confirmatory factor analysis whether scale invariant across sex, type school, educational level variables. sample consisted 2,148 students II, higher education public private schools....

Meta-analyses show that the correlations of approaches to learning with academic achievement are low. However, only one study has controlled these in presence intelligence. Our investigates incremental validity surface and deep taking cognitive abilities as control applied multiple linear regression CART tree method a diversified sample high school students from Minas Gerais, Brazil. We found prediction 20.37% 35.20% outcome variance, through CART, respectively. results although more...

Information from the population should be directly transposed to individual level only under strict conditions of stationarity and homogeneity. In general, psychological phenomena are neither stationary nor homogeneous. Furthermore, parameters must estimated. The usual techniques estimating rarely attainable. This occurs because in order get valid estimates parameters, many occasions same test performed. this article, we propose an approach (simerg algorithm) estimate individual, so that a...

The CART algorithm has been extensively applied in predictive studies, however, researchers argue that produces variable selection bias.This bias is reflected the preference of selecting predictors with large numbers cutpoints.Considering this problem, article compares to an unbiased (CTREE), relation their power.Both algorithms were 2011 National Exam High School Education, which includes many categorical a number categories, could produce bias.A CTREE tree and generated, both 16 leaves,...

Any quantitative method is shaped by certain rules or assumptions which constitute its own rationale. It not chance that these determine the conditions and constraints permit evidence to be constructed. In this article, we argue why Regression Tree Method’s rationale more suitable than General Linear Model analyze complex educational datasets. Furthermore, apply CART algorithm of Method Multiple in a model with 53 predictors, taking as outcome students’ scores reading 2011’s edition National...

Students’ approaches to learning are an important tool used by researchers in the field of educational psychology. The theory assumes that students have two basic ways approaching objects knowledge, a deep one and superficial one. Both reasonable predictors academic performance. study process requires instruments be applied students, nevertheless, exclusive use self-report has substantially hindered adequate this process. Laboratory for Research on Cognitive Architecture (LAICO) at Federal...

The area of self-regulated learning integrates the fields metacognition and self-regulation assumes that student is an active processor information capable self-regulating his by putting together cognitive, metacognitive, motivational components. Motivated Strategies for Learning Questionnaire (MSLQ) a benchmark measurement learning. However, field study does not show adequate evidence its structural validity. vast majority studies involving this question present serious methodological...

This preprint shows initial evidence that the complexity cost criterion may not be a good approach in regression tree method

<p style="text-align:justify">The field of studies in metacognition points to some limitations the way construct has traditionally been measured and shows a near absence performance-based tests. The Meta-Text is test recently created assess components cognition regulation: planning, monitoring, judgment. This study presents first evidence on structural validity Meta-Text, by analyzing its dimensionality reliability sample 655 Honduran university students. Different models were tested,...

Studies on prediction recognize the presence of a large number predictors school-academic performance, as well existence well-established predictors, nevertheless some psychoeducational variables remain little explored, even though their great potential. One these is reflective thinking. The Laboratory for Research Cognitive Architecture (LAICO) at Federal University Minas Gerais brought forth contribution to This development Interest Scale Reflective Thinking which measures interest people...

The theory of students’ approaches to learning makes direct contributions the teaching-learning process. For instance, it provides a conceptual repertoire that helps improve teacher’s pedagogical practice and student’s self-assessment their own learning. Despite its contributions, field has an important limitation: measurement constructs is carried out exclusively by self-report tests. Although these measures have brought advances field, permeated biases impair quality measurement....

SLAT-Thinking is the only test that evaluates and distinguishes stages of approaches through performance. Although shows evidence internal validity, its external validity has not yet been examined. In this paper we study predictive incremental validities SLAT-Thinking. Two models were tested. The predictors inductive reasoning, Learning Approaches Scale (EABAP) approaches. outcome was Brazilian large-scale exam students finish secondary education. both models, superficial approach main...

The study of epidemiological models are important because they help researchers to understand and propose possible strategies combat any epidemic virus. Most the research in those models, however, focuses on response variable, modeling how it varies as a function parameters. In this paper, other hand, we focus explanatory variable ”time,” examining critical points logistic model curve. These are: maximum acceleration point(map), inflection point(ip), deceleration point(mdp), asymptotic...

Abstract Epidemiological models have become a very important tool in understanding an epidemic’s development, mainly because they help researchers find more efficient strategies their fight against its spread. Several been proposed up to now: some use fractional calculus solve differential equations while others applications from other areas such as predatorprey models. The SIR and SEIR models, among others, focus on the variable response epidemiological parameters basic reproduction number...

Latent class analysis (LCA) is used to analyse data about performance of students. The original variables in the set are course grade and approval status. However, those were not directly, instead four new calculated from previous two, which much more informative student performance. Coupled with another variable, say afirmative action, results give light an understanding divided by yes or no. Besides showing it also shown how changing some suitable transformation ones gives reliable...

Abstract Epidemiological models have become a very important tool in understanding an epidemic development, mainly because they help researchers finding good and new strategies their fight against its spread. Several been proposed up to now, some are mathematical others apply from other areas. The SIR SEIR among others, focus on the variable response epidemiological parameters as basic reproduction number (R 0 ) infection rate per unit of time, nevertheless do not ‘time’. We propose use time...

We study estimation of the parameter p a r.v. x ∼ Bin(p,n) using prior hypothesis that equals prepecified value Po when we entertain possibility is not right p, which then could be any in (0.1). apply notions Hodges and Lehmann (1952), Bickel (1984) Berger (1982) to obtain (biased) estimators do well under = po at small price maximum risk. A number examples comparisons are discussed end paper.