NEWS.md

In reproducer: Reproduce Statistical Analyses and Meta-Analyses

CHANGES IN reproducer VERSION 0.5.3

UPDATED FEATURES

• Updated functions calculateLargeSampleRandomizedDesignEffectSizes, NP2GMetaAnalysisSimulation, NP4GMetaAnalysisSimulation (fixed errors)
• Exported functions calculatePhat, calculateCliffd
• Corrected several typos in the documentation of the package

REMOVED FEATURES

• Depreciated functions calculate2GMdMRE, calculate4GMdMRE

CHANGES IN reproducer VERSION 0.5.2

UPDATED FEATURES

• Updated function calculateCliffd (added export)
• Updated functions calculate2GMdMRE, calculate4GMdMRE (fixed CentralPHatMdMRE calculation)

CHANGES IN reproducer VERSION 0.5.1

UPDATED FEATURES

• Updated functions calculate2GMdMRE, calculate4GMdMRE
• PHatMdMRE replaced by CentralPHatMdMRE

CHANGES IN reproducer VERSION 0.5.0

UPDATED FEATURES

• Corrected cross-over data set descriptions
• Updated functions simulateRandomizedBlockDesignEffectSizes, NP4GroupMetaAnalysisSimulation (now NP4GMetaAnalysisSimulation), RandomizedBlockDesignEffectSizes, percentageInaccuracyOfLargeSampleVarianceApproximation
• Changed function names: NP4GroupMetaAnalysisSimulation now NP4GMetaAnalysisSimulation, NP2GroupMetaAnalysisSimulation now NP2GMetaAnalysisSimulation, Kendalltaupb now calculateKendalltaupb, CalculateTheoreticalEffectSizes now calculatePopulationStatistics
• New functions: AnalyseResiduals calc.a calc.b calcCliffdConfidenceIntervals calcEffectSizeConfidenceIntervals calcPHatConfidenceIntervals calculate2GMdMRE calculate4GMdMRE calculateCliffd calculateLargeSampleRandomizedDesignEffectSizes calculateLargeSampleRandomizedBlockDesignEffectSizes calculateNullESAccuracy CatchError checkIfValidDummyVariable Cliffd.test crossoverResidualAnalysis doLM metaanalyseSmallSampleSizeExperiments NP2GMetaAnalysisSimulation NP4GMetaAnalysisSimulation PHat.test simulate2GExperimentData simulate4GExperimentData testfunctionParameterChecks

CHANGES IN reproducer VERSION 0.4.2

UPDATED FEATURES

• Corrected typo in @export of simulateRandomizedDesignEffectSizes
• Corrected typo - lack of ‘@‘ in title and description of RandomizedBlocksExperimentSimulations

CHANGES IN reproducer VERSION 0.4.1

NEW FEATURES

• New functions including computational procedures used to reproduce the main findings in a joint paper (planned to be submitted): Barbara Kitchenham, Lech Madeyski, "Non-parametric Effect Sizes and Their Meta-analysis": varStandardizedEffectSize, RandomizedBlocksAnalysis, Kendalltaupb, Cliffd, calculatePhat, Calc4GroupNPStats, LaplaceDist, simulateRandomizedDesignEffectSizes, RandomExperimentSimulations, simulateRandomizedBlockDesignEffectSizes, RandomizedBlocksExperimentSimulations, NP4GroupMetaAnalysisSimulation, NP2GroupMetaAnalysisSimulation, MetaAnalysisSimulations, CalculateTheoreticalEffectSizes, RandomizedDesignEffectSizes, RandomizedBlockDesignEffectSizes

CHANGES IN reproducer VERSION 0.4.0

NEW FEATURES

• Data set: KitchenhamEtAl.CorrelationsAmongParticipants.Madeyski10, KitchenhamEtAl.CorrelationsAmongParticipants.Scanniello17TOSEM, KitchenhamEtAl.CorrelationsAmongParticipants.Ricca10TSE, KitchenhamEtAl.CorrelationsAmongParticipants.Romano18ESEM, KitchenhamEtAl.CorrelationsAmongParticipants.Scanniello14JVLC, KitchenhamEtAl.CorrelationsAmongParticipants.Reggio15SSM, KitchenhamEtAl.CorrelationsAmongParticipants.Gravino15JVLC, KitchenhamEtAl.CorrelationsAmongParticipants.Ricca14TOSEM, KitchenhamEtAl.CorrelationsAmongParticipants.Scanniello14EASE, KitchenhamEtAl.CorrelationsAmongParticipants.Abrahao13TSE, KitchenhamEtAl.CorrelationsAmongParticipants.Torchiano17JVLC, KitchenhamEtAl.CorrelationsAmongParticipants.Scanniello15EMSE, KitchenhamEtAl.CorrelationsAmongParticipants.Scanniello14TOSEM,

• New functions including computational procedures used to reproduce the main findings in a joint paper (planned to be submitted): Barbara Kitchenham, Lech Madeyski, Giuseppe Scanniello and Carmine Gravino, "The Importance of the Correlation in Crossover Experiments": CalculateRLevel1, ExtractGroupSizeData, ConstructLevel1ExperimentRData, ExtractExperimentData, CalculateLevel2ExperimentRData, ExtractSummaryStatisticsRandomizedExp, calculateBasicStatistics, calculateGroupSummaryStatistics, rSimulations

CHANGES IN reproducer VERSION 0.3.1

NEW FEATURES

• Data set: MadeyskiLewowski.IndustryRelevantGitHubJavaProjects20191022 - over 15\% of entries present in this data set is not present in the previous data set MadeyskiLewowski.IndustryRelevantGitHubJavaProjects20190324 due to moved time windows for the project creation and last push dates.

UPDATED FEATURES

• Updated function searchForIndustryRelevantGitHubProjects - now supports flexible creation date and last push thresholds (enabling the script to better support researchers interested in gathering evolving data sets).

CHANGES IN reproducer VERSION 0.3.0

NEW FEATURES

• Functions including computational procedures used to reproduce the main findings, e.g., tables, in a joint paper: Barbara Kitchenham, Lech Madeyski, Pearl Brereton, "Meta-analysis for Families of Experiments in Software Engineering: A Systematic Review and Reproducibility and Validity Assessment": transformHgtoZr,
• Function to search for industry relevant software projects available from GitHub related to a report: Lech Madeyski, “Training data preparation method,” tech. rep., code quest (research project NCBiR POIR.01.01.01-00-0792/16), 2019; as well as a paper: Tomasz Lewowski and Lech Madeyski, "Creating evolving project data sets in software engineering", 2019: searchForIndustryRelevantGitHubProjects
• Data set: MadeyskiLewowski.IndustryRelevantGitHubJavaProjects20190324

UPDATED FEATURES

• Updated function reproduceTablesOfPaperMetaAnalysisForFamiliesOfExperiments

CHANGES IN reproducer VERSION 0.2.1

UPDATED FEATURES

• Fixed version of the package in README
• Added warning to the ExtractMAStatistics function: it works with metafor version 2.0-0, but changes to metafor's method of providing access to its individual results may introduce errors into the function.

CHANGES IN reproducer VERSION 0.2.0

NEW FEATURES

• functions including computational procedures used to reproduce the main findings, e.g., tables, in a joint paper: Barbara Kitchenham, Lech Madeyski, Pearl Brereton, "Meta-analysis for Families of Experiments: A Systematic Review and Reproducibility Assessment": calculateSmallSampleSizeAdjustment, constructEffectSizes, transformRtoZr, transformZrtoR, transformHgtoR, calculateHg, transformRtoHg, transformZrtoHgapprox, transformZrtoHg, PrepareForMetaAnalysisGtoR, ExtractMAStatistics, aggregateIndividualDocumentStatistics, reproduceTablesOfPaperMetaAnalysisForFamiliesOfExperiments.
• unit tests
• data sets used in the aforementioned paper: KitchenhamMadeyskiBrereton.MetaAnalysisReportedResults, KitchenhamMadeyskiBrereton.ABBAMetaAnalysisReportedResults, KitchenhamMadeyskiBrereton.ReportedEffectSizes, KitchenhamMadeyskiBrereton.ABBAReportedEffectSizes KitchenhamMadeyskiBrereton.ExpData, and KitchenhamMadeyskiBrereton.DocData

UPDATED FEATURES

CHANGES IN reproducer VERSION 0.1.9

NEW FEATURES

• added unit tests and assertions (run-time tests)

UPDATED FEATURES

• updated data set MadeyskiKitchenham.EUBASdata and functions getEffectSizesABBA, effectSizeCI

CHANGES IN reproducer VERSION 0.1.8

NEW FEATURES

• added function getTheoreticalEffectSizeVariancesABBA

UPDATED FEATURES

• updated functions getSimulationData, plotOutcomesForIndividualsInEachSequenceGroup, getEffectSizesABBA, effectSizeCI

CHANGES IN reproducer VERSION 0.1.7

NEW FEATURES

• added function effectSizeCI to calculate 95% Confidence Intervals (CI) on Standardised Effect Sizes (d) for cross-over repeated-measures designs

REMOVED FEATURES

• removed field dIG.Var.CurtinAve from the data frame returned by the reproduceSimulationResultsBasedOn500Reps1000Obs function (we agreed to write joint paper with Dr Curtin describing corrections to his equations to calculate effect size variances for continuous outcomes of cross-over clinical trials)

CHANGES IN reproducer VERSION 0.1.6

NEW FEATURES

• added functions to reproduce simulation data, to calculate effect sizes and their variance discussed in a paper "Effect Sizes and their Variance for AB/BA Crossover Design Studies" by Lech Madeyski and Barbara Kitchenham.
  • getSimulationData
  • plotOutcomesForIndividualsInEachSequenceGroup
  • getEffectSizesABBA
  • getEffectSizesABBAIgnoringPeriodEffect
  • reproduceSimulationResultsBasedOn500Reps1000Obs
  • percentageInaccuracyOfLargeSampleVarianceApproximation
  • proportionOfSignificantTValuesUsingCorrectAnalysis
  • proportionOfSignificantTValuesUsingIncorrectAnalysis
• new data set:
  • KitchenhamMadeyski.SimulatedCrossoverDataSets backed by functions (varianceSimulation, getSimulatedCrossoverDataSets) to reproduce the data set.

REMOVED FEATURES

• removed function:
  • cloudOfWords

CHANGES IN reproducer VERSION 0.1.5

BUG FIXES

• corrected typo in a surname of one of the data contributors
• corrected the name of the university in Hong Kong
• corrected the boxplotAndDensityCurveOnHistogram function
• updated references to research papers supported by the package

CHANGES IN reproducer VERSION 0.1.4

NEW FEATURES

• new data sets:
  • KitchenhamMadeyskiBudgen16.FINNISH
  • KitchenhamMadeyskiBudgen16.PolishSubjects
  • KitchenhamMadeyskiBudgen16.SubjectData
  • KitchenhamMadeyskiBudgen16.PolishData
  • KitchenhamMadeyskiBudgen16.DiffInDiffData
  • KitchenhamMadeyskiBudgen16.COCOMO
• updated to be compatible with the ggplot2 2.0.0
• minor visual changes in figures produced by the following functions:
  • densityCurveOnHistogram
  • boxplotHV
  • boxplotAndDensityCurveOnHistogram

CHANGES IN reproducer VERSION 0.1.3

NEW FEATURES

• added general functions
  • printXTable
  • cloudOfWords
• added functions to reproduce all of the tables, figures and outputs in the paper by Lech Madeyski and Barbara Kitchenham on how variations in experimental designs impact the construction of comparable effect sizes for meta-analysis:
  • reproduceForestPlotRandomEffects
  • reproduceMixedEffectsAnalysisWithEstimatedVarianceAndExperimentalDesignModerator
  • reproduceMixedEffectsAnalysisWithExperimentalDesignModerator
  • reproduceMixedEffectsForestPlotWithExperimentalDesignModerator
  • reproduceTableWithEffectSizesBasedOnMeanDifferences
  • reproduceTableWithPossibleModeratingFactors
  • reproduceTableWithSourceDataByCiolkowski
• added data sets:
  • Ciolkowski09ESEM.MetaAnalysis.PBRvsCBRorAR
  • MadeyskiKitchenham.MetaAnalysis.PBRvsCBRorAR

BUG FIXES

• corrected Madeyski15EISEJ.StudProjects$STUD data set

CHANGES IN reproducer VERSION 0.1.2

NEW FEATURES

• The first published version of reproducer. It includes data sets:
  • Madeyski15SQJ.NDC
  • Madeyski15EISEJ.OpenProjects
  • Madeyski15EISEJ.PropProjects
  • Madeyski15EISEJ.StudProjects and functions (for importing data, visualization and descriptive analyses):
  • readExcelSheet
  • densityCurveOnHistogram
  • boxplotHV
  • boxplotAndDensityCurveOnHistogram

See the package homepage (https://madeyski.e-informatyka.pl/reproducible-research/) for documentation and examples.

reproducer documentation built on Oct. 18, 2023, 5:10 p.m.