The JASP Data Library
The JASP TeamThe JASP Data LibraryVersion 1JASP Publishing
Copyright 2019 E.-J. Wagenmakerspublished by jasp publishingtufte-latex.googlecode.comLicensed under the Apache License, Version 2.0 (the “License”); you may not use this file except in compliance with the License. You may obtain a copy of the License at http://www.apache.org/licenses/LICENSE-2.0. Unless required by applicable law or agreed to in writing, software distributed under theLicense is distributed on an “as is” basis, without warranties or conditions of any kind, eitherexpress or implied. See the License for the specific language governing permissions and limitations underthe License.This printing, March 2019
ContentsContentsPrefacePART I37DESCRIPTIVE STATISTICS1 Sleep: Descriptive Statistics192 Fear of Statistics: Reliability AnalysisBibliography35List of External Figures3727
5“You don’t understand anything until youlearn it more than one way.”Marvin Minsky
PrefaceThis is the documentation for the examples contained in the JASPData Library. The data sets are available from JASP both in .jaspformat and in .csv format. This documentation will grow over time,and therefore we have added a version number to the book title.Structure and ContentsThe structure of this documentation follows the layout of the JASPribbon: the first parts deal with Descriptives, T-tests, ANOVA, Regression, Frequencies, and Factor Analysis, whereas the remaining partsconcern various JASP modules, namely Meta-Analysis, NetworkAnalysis, and Structural Equation Modeling.The example data sets come from various sources. For educationalpurposes we have included the material from two popular statisticaltextbooks, Field (2017) and Moore et al. (2012). This means it is noweasy for teachers and students to use JASP and analyze the classic‘Looks or Personality’ example from Andy Field, or the ‘CollegeSuccess’ example from Moore, McCaib, and Craig. We also added theoriginal example data sets that JASP users may be familiar with (everwonder what the ‘Kitchen Rolls’ data set was all about?). A numberof publicly available data sets have been added as well.CopyrightMost data sets contained in the JASP Data Library are already publicly available because they have been presented in academic articlesand books. The legal consensus is that such data are facts and therefore do not fall under copyright protection. An exception concernsdata that have not been collected but constructed in order to servean educational purpose, as is the case for most data sets from AndyField’s popular course books. For these data sets, the source is acknowledged as follows:“This data set comes from Field, A. P. (2017). Discovering Statistics Using IBM SPSS Statistics (5th ed.). London: Sage. The dataFigure 1: JASP: A freshway to do statistics. Download JASP for free at http://jasp-stats.org.
8set was constructed by Andy Field who therefore owns the copyright. Andy Field generously agreed that we can include the dataset in the JASP data library. This data set is also publicly available on the website that accompanies Andy Field’s book: https://edge.sagepub.com/field5e. Without Andy Field‘s explicit consent,this data set may not be distributed for commercial purposes, this dataset may not be edited, and this data set may not be presented withoutacknowledging its source (i.e., the terms of a CC BY-NC-ND -nd/3.0/). ”The Need for Documentation GuidelinesOur experience with the analysis of publicly available data suggeststhat there is an urgent need for standard guidelines that encouragesystematic and comprehensive documentation. Without such guidelines, the analysis often becomes a frustrating exercise in encryption.1Oftentimes, the data could not have been properly analyzed withoutassistance from the original authors. This undesirable state of affairsis why we have developed the JASP Data Documentation Formatdescribed below. This format provides a list of minimal requirementsfor how data ought to be documented, one that may need to be expanded for particular data sets (such as those from neuroscience).The proposed format is concrete and focuses purely on documentation (see the FAIR guidelines2 for complementary advice concerningdata storage).For empirical support, see for instanceHardwicke et al. (2018).12Wilkinson et al. (2016).3JDDF templates are available online atThe JASP Data Documentation FormatEach chapter in this book documents a single data set using theJASP Data Documentation Format (JDDF).3 The JDDF contains thefollowing elements:I. Description. A brief summary of the data set in order to providecontext. Example:This data set, Glasgow Norms, provides a set of normative ratings of 5, 553 English words on nine psycholinguistic dimensions.Specifically:“The Glasgow Norms are a set of normative ratings for 5,553English words on 9 psycholinguistic dimensions: arousal,valence, dominance, concreteness, imageability, familiarity, ageof acquisition, semantic size, and gender association (.) forany given subset of words, the same participants providedratings across all 9 dimensions (32 participants/word, onaverage).” (Scott et al. 2017, p. 2).II. Purpose. A brief statement of the purpose that the data may serve.Example:https://osf.io/bjmrg/
9The purpose of the original study was to develop a substantial set ofstandardized, freely available, psycholinguistic materials. Scott et al.(2017) state that“Overall, the Glasgow Norms provide a valuable resource,in particular, for researchers investigating the role of wordrecognition in language comprehension” (Scott et al. 2017,p. 2).Here we use this data set to demonstrate Bayesian Pearson’scorrelation. Specifically, we will estimate how the psycholinguisticdimensions relate to each other. For example, are words that arerated higher on the Dominance dimension rated as more masculineor feminine? Are words associated with higher arousal rated morepositive or negative?III. Data Screenshot. A visual impression of the data structure, showing at least some of the columns and rows in which the data havebeen organized. An example screenshot is shown in Figure 2. Anexample of the text that could accompany the screenshot:The Glasgow Norms data set (Scott et al. 2017) consists of 5, 553rows and 29 columns. Each row corresponds to a specific word, andthe columns relate to various lexical properties.IV. Variables. A description for each of the variables in the data set.At a minimum, the description features (a) the variable name; (b)the variable’s measurement level and the values that the variablecan take on; (c) a verbal summary of the variable. An example forthe first five variables in the Glasgow Norms data set: word– Type: Categorical– Description: The word whose lexical properties are listed. length– Type: Continuous– Description: Word length in number of letters. M AROU– Type: Continuous– Description: A word’s mean arousal value (averaged acrossraters). Each rater judged the extent to which a word hasemotional impact in terms of internal activation on a 9-pointscale (‘1’ very unarousing, ‘9’ very arousing). SD AROU– Type: ContinuousIn this book, we use three types ofvariables: Categorical, Ordinal, andContinuous. Even discrete variables,although not only strictly continuous,are categorized as ’Continuous’ - thisis because the analyses options in JASPare the same for both types of variables.
10– Description: The standard deviation (across raters) of a word’sarousal value. Each rater judged the extent to which a wordhas emotional impact in terms of internal activation on a9-point scale (‘1’ very unarousing, ‘9’ very arousing). N AROU– Type: Continuous– Description: The number of raters who judged a word’sarousal value. Each rater judged the extent to which a wordhas emotional impact in terms of internal activation on a9-point scale (‘1’ very unarousing, ‘9’ very arousing).V. Source. A description of where the data were obtained, and wherethe data can be accessed online. Example:Scott, G. G., Keitel, A., Becirspahic, M., O’Donnell, P. J., and Sereno,S. C. (2017). The Glasgow norms: Ratings of 5,500 words on 9 scales.PsyArXiv, https://psyarxiv.com/akzyx/. The data set is availableas Supplementary Materials.Figure 2: Screenshot of the first18 rows and 8 columns of theGlasgow Norms data (Scottet al. 2017).
11VI. Analysis code. Anything that allows a third party to reproduce aresult. This can be R code, Stan code, or a .jasp file, for instance.We strongly recommend against using commercial software foranalysis, as the interested third party might be unwilling or unableto buy an expensive license (e.g., to Matlab, JMP, SAS, Stata, orSPSS) in order to reproduce the result. Example:An annotated JASP file is available in the JASP Data Library and onthe OSF at https://osf.io/ypwk5/.VII. Example analysis. A demonstration that the provided code, whenapplied to the provided data, actually produces a result. Example:As a first step, we will inspect the correlations between the ninepsycholinguistic dimensions. Open the relevant analysis windowunder the Regression Bayesian Correlation Matrix optionfrom the Common menu. A screenshot of the relevant JASP inputpanel is shown in Figure 3. Drag all nine dimensions into thevariable selection window.The size of the data set and the number of tests involved meansthat we have to wait a little before JASP produces the output shownin Figure 4. By default, JASP returns the classical estimate of thePearson correlation coefficient r for each of the pairwise associationsbetween the variables. For each association, JASP also presents theBayes factor BF10 that quantifies the extent to which the alternativehypothesis H1 outpredicts the null hypothesis H0 (Wagenmakerset al. 2016). The null hypothesis assumes that the true correlationis absent, that is, the latent correlation coefficient ρ equals 0. Thealternative hypothesis assumes that the uncertainty about ρ is givenby a stretched beta distribution which is symmetric around ρ 0;when the width is set to 1, the prior is uniform and each valueof ρ is deemed equally likely a priori (Ly et al. 2018). Note thatfor most of the correlations we obtain overwhelming evidence forthe alternative hypothesis. Apart from three correlations (‘AOA’‘AROU’, ‘SIZE’-‘FAM’, ‘CNC’-‘DOM’), all Bayes factors show strongevidence for the existence of a correlation between the relevantdimensions. The large sample size makes it relatively easy to obtainfavorable evidence for the alternative hypothesis even when the sizeof the correlations is small.To learn about the uncertainty of the correlation estimates underH1 , we can ask JASP to produce credible intervals. Figure 4 showsthat JASP provides, for each pairwise association, both the lowerand the upper bound of the central 95% credible interval. It isevident that the magnitude of the correlations varies considerably;whereas some correlations are as large as r .91 (‘IMAG’-‘CNC’),others are as small as .06 (‘CNC’-‘VAL’). Some dimensions actuallyseem uncorrelated; for example, Arousal (‘AROU’) and Age ofAquisition (‘AOA’) yield a Pearson’s r 0.000, a 95% CI (givenH1 ) of [ 0.026, 0.026], and a BF10 0.017, which means that H0outpredicted H1 by a factor of 1/0.017 58.82.
12Figure 3: Screenshot of theJASP input panel for the analysis of the Glasgow Normsdata (Scott et al. 2017) using theBayesian Pearson’s correlation.
13Figure 4: JASP output for theGlasgow Norms data (Scottet al. 2017): Bayesian correlation matrix. Apart from threecases, all Bayes factors showstrong support for the existenceof correlations between thedimensions.
14To gain more insight into the correlation patterns and check thevalidity of the analysis, it is advisable to plot the data. This canbe done by ticking the check boxes under the Plots section in theanalysis menu (see Figure 3). The produced output can be seen inFigure 5. The lower half of the correlation plot shows the posteriordistributions of the Bayesian correlation coefficients. We can seethat all posterior distributions are highly peaked, indicating greatcertainty about the magnitude of the correlations. The diagonal ofthe plot shows histograms and density functions of the distributionfor the single variables. For most of the dimensions, there are bothvery high-rated and very low-rated words so that the whole scaleis used. However, for most dimensions, the peak of the histogramfalls in the middle of the scale. The upper part of the plot shows onescatter plot for each correlation, displaying the joint distribution ofthe two pertinent variables.VIII. Want to know more? Any additional information of interest.Example:3 Ly, A., Marsman, M., and Wagenmakers, E.-J. (2018). Analytic posteriorsfor Pearson’s correlation coefficient. Statistica Neerlandica, 72:4–133 Scott, G. G., Keitel, A., Becirspahic, M., O’Donnell, P. J., and Sereno,S. C. (2017). The Glasgow norms: Ratings of 5,500 words on 9 scales.PsyArXiv3 Wagenmakers, E.-J., Morey, R. D., and Lee, M. D. (2016). Bayesianbenefits for the pragmatic researcher. Current Directions in PsychologicalScience, 25:169–176ContributorsThis data library is a product of the JASP team. Most of the work wasdone by Šimon Kucharský and Eric-Jan Wagenmakers with assistancefrom Angelika Stefan, Lotte Kehrer, and Sophia Crüwell.AcknowledgmentsThis documentation contains numerous illustrations. Many havebeen taken from Wikepedia and fall in their ‘fair use’ category. Headshots have occasionally been taken from publicly accessible websites.Other illustrations we have created ourselves, and most screenshotscome from JASP. Detailed information concerning the external figuresis provided in the final chapter. We are indebted to the creators ofthe Tufte LATEX style files (and in particular to Kevin Godby), to theOverleaf editing system, and to Wikipedia. We also wish to thankthose who have proofread this library and recommended importantimprovements.
15Figure 5: JASP output for theGlasgow Norms data (Scottet al. 2017): Bayesian correlations plot. The plot shows thedistributions in scatterplots(upper part), posterior distributions of Pearson correlationcoefficients ρ (lower part), anddistributions of individualvariables (diagonal).
Part IDescriptive Statistics
1Sleep: Descriptive StatisticsDescriptionThis data set, Sleep, provides the number of additional hours thateach of ten patients slept after having been administered two ’soporific drugs’ (i.e., sleeping pills).Did You Know?The data set reported in Cushny and Peebles (1905) has made an interesting mark in the history of statistics. Under his pen name Student,William Sealy Gosset used this data set in his famous article The probable error of the mean to illustrate the test that later became knownas Student’s t-test. Later on, the data set was discussed by anotherfamous statistician, Ronald Aylmer Fisher, in his classic book StatisticalMethods for Research Workers (1925). However, Gosset made a transcription error in the column labels, an error which Fisher followed, whichmeans that two of the most influential statisticians of the 20th centurymisinterpreted the data set! For details see Senn (2017).1.1PurposeThe original purpose of the study by Cushny and Peebles (1905) wasto compare the effectiveness of two ‘soporific drugs’ in a withinsubjects design with 10 patients.Here we use this data set to demonstrate how to generate descriptive statistics and plots for different groups in JASP.1.2Data ScreenshotPart of the original data set is depicted in Figure 1.2. The Sleep dataset (Cushny and Peebles 1905) consists of 20 rows and 3 columns.Each participant has been administered both drugs and thereforeoccupies two rows.Figure 1.1: Sleep. Image byJackman Chiu under CC0license.
20the jasp data library1.3VariablesFigure 1.2: Screenshot of thefirst five rows of the Sleep data(Cushny and Peebles 1905). extra– Type: Continuous– Description: Increase in hours of sleep relative to a control drug. group– Type: Categorical– Description: Type of drug (1 first drug, 2 second drug). ID– Type: Categorical– Description: Number that indexes the participant.1.4SourceCushny, A. R. and Peebles, A. R. (1905). The action of optical isomers.The Journal of Physiology, 32:501–5101.5Analysis CodeFigure 1.3: Data set includedin R: An annotated JASP file is available in the JASP Data Library and onthe OSF at https://osf.io/6stjy/.1.6Example AnalysisLet’s take a look at some descriptive statistics of the sleep increasefor both drugs separately. To produce descriptive statistics, openthe relevant analysis options under Descriptives DescriptiveStatistics in the Common menu. A screenshot of the relevant JASPbase/html/sleep.html
sleep: descriptive statistics21input panel is shown in Figure 1.4. First, we drag the variable ‘extra’into the Variables window. The Descriptives input panel allows usto split the data set by separate categories. We will use this to splitthe output by the ‘group’ variable, which indicates the different drugconditions. We can drag the variable ‘group’ into the Split box andreceive a table with traditional descriptive statistics separately foreach drug condition. We can also generate descriptive plots. Go tothe Plots section and select Display boxplots. To have all possibleinformation included in the graph, check the Color, Violin Element,and Jitter Element box. The Boxplot Element is checked by default.This setup will print all three elements on top of each other for thetwo groups separately, as shown in Figure 1.5.All three elements shown in Figure 1.5 provide complementaryinformation about the data at hand: the Jitter Element reveals theindividual data points (note that the jitter along the x-axis allowsFigure 1.4: Screenshot of theJASP input panel for descriptivestatistics of the Sleep data set(Cushny and Peebles 1905).
22the jasp data libraryus to differentiate between patients that have similar increase insleep). The Violin Element approximates the sample distribution(offering insight into potential outliers and normality – includingskew, kurtosis, and unimodality). The Boxplot Element summarizesthe sample distribution in more classical fashion – showing its range,interquartile range, and median.However, the data is structured in the so-called ‘long-format’.This means that each row corresponds to one observation and notnecessarily to one participant. In fact, we have two observations foreach participant (as all of them had been administered both drugson different occasions) and so each participant occupies two rows.The nature of the data suggest that we would also like to know thecorrelations between the different conditions. In order to inspect thecorrelations, we need to reshape the data to a ‘wide-format’ such thateach row corresponds to an individual participant and the increase insleep occupies two columns, one for each of the two drug conditions.Luckily, JASP allows us to reshape the data in any spreadsheetprogram we like. Double-click on any cell in the data viewer to openthe data set in our predefined default spreadsheet editor, for instanceLibreOffice ‘Calc’ (you can choose a different program using thePreferences option available under the hamburger menu icon on thetop right of the JASP screen). This way we can restructure the datainto the desired format. The reshaped data are shown in Figure 1.7.Figure 1.5: JASP output for theSleep data (Cushny and Peebles1905): Colored boxplots withviolins and jittered data areshown separately for the twodrug conditions.Figure 1.6: Click on the hamburger menu icon to openPreferences option for JASP.
sleep: descriptive statistics23Figure 1.7: Restructured Sleepdata (Cushny and Peebles 1905):Each participant now occupiesone row, whereas the amountof extra sleep compared to ano drug condition is in twoseparate columns indicating thetwo diff
LICENSE-2.0. Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an “as is” basis, without warranties or conditions of any kind, either express or implied. See the License for the specific language governing permissions and limitations under the License. This printing, March 2019
May 02, 2018 · D. Program Evaluation ͟The organization has provided a description of the framework for how each program will be evaluated. The framework should include all the elements below: ͟The evaluation methods are cost-effective for the organization ͟Quantitative and qualitative data is being collected (at Basics tier, data collection must have begun)
Silat is a combative art of self-defense and survival rooted from Matay archipelago. It was traced at thé early of Langkasuka Kingdom (2nd century CE) till thé reign of Melaka (Malaysia) Sultanate era (13th century). Silat has now evolved to become part of social culture and tradition with thé appearance of a fine physical and spiritual .
On an exceptional basis, Member States may request UNESCO to provide thé candidates with access to thé platform so they can complète thé form by themselves. Thèse requests must be addressed to esd rize unesco. or by 15 A ril 2021 UNESCO will provide thé nomineewith accessto thé platform via their émail address.
̶The leading indicator of employee engagement is based on the quality of the relationship between employee and supervisor Empower your managers! ̶Help them understand the impact on the organization ̶Share important changes, plan options, tasks, and deadlines ̶Provide key messages and talking points ̶Prepare them to answer employee questions
Dr. Sunita Bharatwal** Dr. Pawan Garga*** Abstract Customer satisfaction is derived from thè functionalities and values, a product or Service can provide. The current study aims to segregate thè dimensions of ordine Service quality and gather insights on its impact on web shopping. The trends of purchases have
4 P a g e JASP 0.14 - Dr Mark Goss-Sampson In the Results Preferences section users can: Set JASP to return exact p values i.e. P 0.00087 rather than P .001 Fix the number of decimals for data in tables – makes tables easier to read/publish
Chính Văn.- Còn đức Thế tôn thì tuệ giác cực kỳ trong sạch 8: hiện hành bất nhị 9, đạt đến vô tướng 10, đứng vào chỗ đứng của các đức Thế tôn 11, thể hiện tính bình đẳng của các Ngài, đến chỗ không còn chướng ngại 12, giáo pháp không thể khuynh đảo, tâm thức không bị cản trở, cái được
Learning Statistics with JASP: A Tutorial for Psychology Students and Other Beginners (Version 1? 2) DanielleNavarro UniversityofNewSouthWales d.navarro@unsw.edu.au
