tag:blogger.com,1999:blog-4736329668300304761Fri, 13 Jul 2018 08:22:03 +0000R-miniRMeta-AnalysisRStudioSPSSStatisticsTeachingAccessibilityBlindConfidence IntervalDistributionEffect SizeFactor StructureParallel ProcessingPeer ReviewPhilosophy of SciencePowerRegressionReliabilityReplicationScale DevelopmentStructural Equation ModelingSubmitting ManuscriptsMy short articles on psychological research: methodology, interesting studies, thoughts, ... anything about psychological research.https://blogonresearch.blogspot.com/noreply@blogger.com (Shu Fai Cheung)Blogger26125tag:blogger.com,1999:blog-4736329668300304761.post-4538528244927495671Wed, 16 May 2018 10:28:00 +00002018-05-16T18:28:38.569+08:00R-mini# A useful c() argument. a <- list(1:2, 3:4) b <- list(5:6, 7:8) c(a, b) # Combines the list to give a list c(a, b, recursive = TRUE) # Combine recursively to give a vector :) https://blogonresearch.blogspot.com/2018/05/r-mini-c-has-arguments-too.htmlnoreply@blogger.com (Shu Fai Cheung)0tag:blogger.com,1999:blog-4736329668300304761.post-8591644648313067574Thu, 03 Aug 2017 12:47:00 +00002017-08-04T05:48:14.603+08:00Parallel ProcessingR Found that future_lapply is better than parLapplyLB in running parallel jobs diverse in speed in Windows. 😊 It took me one afternoon to figure out how to use the future package. I heard about it but no time to learn it before. It's different from parallel because the environment in each worker is not persistent. However, this is a good thing, ensuring a consistent environment. In whathttps://blogonresearch.blogspot.com/2017/08/the-future-package.htmlnoreply@blogger.com (Shu Fai Cheung)0tag:blogger.com,1999:blog-4736329668300304761.post-1445871268349434411Sat, 12 Nov 2016 10:19:00 +00002016-11-13T01:41:31.885+08:00RRStudio Tried R notebook and I like it. The next time I introduce R to new users, I may go directly to R notebook and skip console, or only mention it briefly. Many people have no experience in command line interface, for good reasons. They have a hard time learning and understanding the console. Though lines are still sent to the console when running code in R notebooks, users don't really https://blogonresearch.blogspot.com/2016/11/r-notebook.htmlnoreply@blogger.com (Shu Fai Cheung)0tag:blogger.com,1999:blog-4736329668300304761.post-4912807434949913808Sat, 12 Mar 2016 11:19:00 +00002016-03-12T19:19:12.826+08:00RStructural Equation Modelinghttps://cran.r-project.org/web/packages/lavaan.shiny/ An interesting package! I am going teach lavaan this semester. I will try it first and may introduce it to my student. I still think writing and running script is the most efficient work flow. However, a shiny-based interface may be good for learning and testing. https://blogonresearch.blogspot.com/2016/03/lavaanshiny.htmlnoreply@blogger.com (Shu Fai Cheung)0tag:blogger.com,1999:blog-4736329668300304761.post-8622541050382106914Fri, 11 Dec 2015 22:49:00 +00002015-12-12T06:49:27.515+08:00RRStudioLink: Using Source Windows (@ RStudio Support) A good feature! Some users like one single window with panes. Some users like multiple windows. Some users like being able to choose. I like the interface of GIMP, in which you can switch between single and multiple window modes. I also like the interface of Blender, in which you can work in one window with panes, as in RStudio, or you can work in https://blogonresearch.blogspot.com/2015/12/source-windows-in-rstudio.htmlnoreply@blogger.com (Shu Fai Cheung)0tag:blogger.com,1999:blog-4736329668300304761.post-978192781203845743Wed, 25 Nov 2015 12:33:00 +00002015-11-25T20:33:37.065+08:00AccessibilityBlindRSPSSStatisticsTeachingGodfrey, A. J. R., & Loots, M. T. (2015) Advice from blind teachers on how to teach statistics to blind students. Journal of Statistics Education, 23. Coincidentally, I talked about this with a friend today. I know next to nothing on this topic. I definitely should read this article, to be prepared. Two related articles on the accessibility of statistics software that I found useful. Godfreyhttps://blogonresearch.blogspot.com/2015/11/godfrey-loots-2015-advice-from-blind.htmlnoreply@blogger.com (Shu Fai Cheung)0tag:blogger.com,1999:blog-4736329668300304761.post-1382942193725219888Sat, 14 Nov 2015 12:06:00 +00002015-11-14T20:06:02.697+08:00Confidence IntervalEffect SizePowerReplicationLindsay, D. S. (2015). Replication in psychological science. Psychological Science. doi:10.1177/0956797615616374 A good move. We need to pay attention to power, p-hacking, effect size, confidence interval, distribution (mean and SD are not enough), scatter plot (correlation is not enough), and replication. For many of the issues, we have recommended to pay attention to them for a long time. https://blogonresearch.blogspot.com/2015/11/replication-in-psychological-science.htmlnoreply@blogger.com (Shu Fai Cheung)0tag:blogger.com,1999:blog-4736329668300304761.post-639113062829433456Wed, 11 Nov 2015 11:59:00 +00002015-11-11T20:07:49.904+08:00Meta-AnalysisHedges, L. V., & Olkin, I. (2015) Overlap Between Treatment and Control Distributions as an Effect Size Measure in Experiments. Psychological Methods. http://dx.doi.org/10.1037/met0000042 Feel excited reading this article by Hedges and Olkin, published online last week. It has been 30 years since they published one of the most important books on meta-analysis, Statistical https://blogonresearch.blogspot.com/2015/11/hedges-and-olkin-2015-overlap-between.htmlnoreply@blogger.com (Shu Fai Cheung)0tag:blogger.com,1999:blog-4736329668300304761.post-171114102103022452Fri, 23 Oct 2015 12:58:00 +00002015-10-23T20:58:08.201+08:00Meta-AnalysisArticle: Many Antidepressant Studies Found Tainted by Pharma Company Influence (Roni Jacobson, Scientific American, October 21, 2015) Due to conflict of interests, primary studies may be biased. So we have meta-analyses. And then, also due to conflict of interests, meta-analyses may also be biased. Solution? Maybe transparency and reproducibility.https://blogonresearch.blogspot.com/2015/10/meta-analyses-are-they-immune-to.htmlnoreply@blogger.com (Shu Fai Cheung)0tag:blogger.com,1999:blog-4736329668300304761.post-2941155566661864635Wed, 23 Sep 2015 13:45:00 +00002016-02-20T10:07:03.871+08:00Philosophy of ScienceJust found that this classic is available for free at the Internet Archive! Great! There are PDF and MOBI versions, among other formats. I read some of the chapters several years ago. Yes, it was published nearly 90 years ago and may not be "modern" now. But some of the ideas were still adopted (or misunderstood?) by psychologists in the 21st century (yes, psychologists).https://blogonresearch.blogspot.com/2015/09/book-logic-of-modern-physics-bridgman.htmlnoreply@blogger.com (Shu Fai Cheung)0tag:blogger.com,1999:blog-4736329668300304761.post-8116354945032756254Mon, 24 Nov 2014 15:30:00 +00002014-11-24T23:30:01.226+08:00R-mini Some common ways to find the documentation of a function. To find the documentation of the function mean(): ?mean To search using a reserved word, such as else and if: ?"else" ?"if" To find the documentation of a symbol, for example, the plus symbol +: ?"+" ?"%*%"   To find the documentation files with the string "weighted" somewhere in them: ??weighted To find the documentation fileshttps://blogonresearch.blogspot.com/2014/11/r-mini-help-on-help.htmlnoreply@blogger.com (Shu Fai Cheung)0tag:blogger.com,1999:blog-4736329668300304761.post-8566285659715411962Thu, 20 Nov 2014 15:30:00 +00002014-11-20T23:30:01.263+08:00R-miniSuppose I would like to convert this string, "N=100, NRep=200, K=10, P=5", to a named list, I can use eval() and parse(). This is useful when the arguments for a function is passed as one single string as the command line argument. .sc0 { font-family: 'Courier New'; } .sc3 { font-family: 'Courier New'; color: #8000FF; } .sc5 { font-family: 'Courier New'; color: #FF8000; } .sc6 { https://blogonresearch.blogspot.com/2014/11/r-mini-convert-long-string-to-named-list.htmlnoreply@blogger.com (Shu Fai Cheung)0tag:blogger.com,1999:blog-4736329668300304761.post-2935645462285760176Tue, 18 Nov 2014 15:00:00 +00002014-11-18T23:00:03.904+08:00R-miniSPSSI still occasionally need to read an SPSS data file in R because some of my collaborators use SPSS. I usually use read.spss() from the foreign package: library(foreign) fromSPSS.df <- read.spss(file.choose(), reencode=65001, use.value.labels=TRUE, to.data.frame=TRUE) The output is the data frame fromSPSS.df. file.choose() can be replaced by the file name in quotes if the SPSS data file in the https://blogonresearch.blogspot.com/2014/11/r-mini-read-spss-data-file.htmlnoreply@blogger.com (Shu Fai Cheung)0tag:blogger.com,1999:blog-4736329668300304761.post-3576946786169779715Sun, 16 Nov 2014 15:00:00 +00002014-11-16T23:00:07.291+08:00R-miniAs mentioned in a previous post, "Scatter plot is an important tool for data exploration. However, Likert scales, with only a few discrete possible values, are common in psychology. The usual scatter plot is not good for examining the association between two Likert scale variables. Cross-tabulation table can show the frequencies, but I like using graphs." Bubble plot is introduced in that https://blogonresearch.blogspot.com/2014/11/r-mini-scatter-plots-for-likert-scales_16.htmlnoreply@blogger.com (Shu Fai Cheung)0tag:blogger.com,1999:blog-4736329668300304761.post-1443122440503577967Thu, 13 Nov 2014 15:00:00 +00002014-11-13T23:00:00.392+08:00R-miniScatter plot is an important tool for data exploration. However, Likert scales, with only a few discrete possible values, are common in psychology. The usual scatter plot is not good for examining the association between two Likert scale variables. Cross-tabulation table can show the frequencies, but I like using graphs. One possibility is to use bubble plot, and set the size of each bubble to https://blogonresearch.blogspot.com/2014/11/r-mini-scatter-plots-for-likert-scales.htmlnoreply@blogger.com (Shu Fai Cheung)0tag:blogger.com,1999:blog-4736329668300304761.post-4388201518556437185Sat, 08 Nov 2014 15:30:00 +00002014-11-08T23:30:01.558+08:00R-miniThis article introduces three simple ways to format R code in Blogger. For me, the Notepad++ method is the simplest one, and good enough for short code. However, I need to delete the style for span, and copy the font-family in span to .sc0 and other classes starting with sc, created by the plug-in. But it is simple and good enough for me.https://blogonresearch.blogspot.com/2014/11/r-mini-format-r-code.htmlnoreply@blogger.com (Shu Fai Cheung)0tag:blogger.com,1999:blog-4736329668300304761.post-8532801084369874213Thu, 06 Nov 2014 16:05:00 +00002014-11-07T00:05:48.541+08:00R-mini .sc0 { font-family: 'Courier New'; } .sc2 { font-family: 'Courier New'; font-weight: bold; color: #0000FF; } .sc3 { font-family: 'Courier New'; color: #8000FF; } .sc5 { font-family: 'Courier New'; color: #FF8000; } .sc8 { font-family: 'Courier New'; font-weight: bold; color: #000080; } .sc9 { font-family: 'Courier New'; } This task is very simple. I want to construct a correlation https://blogonresearch.blogspot.com/2014/11/r-mini-from-lower-triangular-elements.htmlnoreply@blogger.com (Shu Fai Cheung)0tag:blogger.com,1999:blog-4736329668300304761.post-8475575217268240809Wed, 05 Nov 2014 12:20:00 +00002014-11-05T20:20:00.442+08:00R-miniFor some reasons, I need to keep several versions of R in my computer, and they need to have their own libraries. The function copy.packages.between.libraries() from the package installr is very useful for this purpose. Right after installing a new version of R, just run the following: copy.packages.between.libraries() The default options are OK for this purpose. All packages will be copied to https://blogonresearch.blogspot.com/2014/11/r-mini-copy-all-packages-to-library-of.htmlnoreply@blogger.com (Shu Fai Cheung)0tag:blogger.com,1999:blog-4736329668300304761.post-6339632169042076196Mon, 03 Nov 2014 12:18:00 +00002014-11-03T20:25:13.475+08:00R-mini .sc0 { font-family: 'Courier New'; } .sc2 { font-family: 'Courier New'; font-weight: bold; color: #0000FF; } .sc3 { font-family: 'Courier New'; color: #8000FF; } .sc8 { font-family: 'Courier New'; font-weight: bold; color: #000080; } .sc9 { font-family: 'Courier New'; } Interestingly, to unload a package, it is not done by something similar to library(), the function used to load a https://blogonresearch.blogspot.com/2014/11/r-mini-unload-package.htmlnoreply@blogger.com (Shu Fai Cheung)0tag:blogger.com,1999:blog-4736329668300304761.post-7577781396264522606Mon, 03 Nov 2014 12:01:00 +00002014-11-03T20:04:15.552+08:00R-mini What are R-mini posts? They are mini notes on R that I would like to keep record and make public for other users. They can be commonly used functions or code snippets. They can be tips for new users. They can be very simple functions most users should already know. They can be tricks for advanced users. They can be something I learned from others and would like to share. They can be anything https://blogonresearch.blogspot.com/2014/11/r-mini-what-are-r-mini-posts.htmlnoreply@blogger.com (Shu Fai Cheung)0tag:blogger.com,1999:blog-4736329668300304761.post-890488149729512581Tue, 15 Jul 2008 13:47:00 +00002008-07-15T21:47:19.269+08:00RegressionArticle: Becker, T. E. (2005). Potential problems in the statistical control of variables in organizational research: A qualitative analysis with recommendations. Organizational Research Methods, 8, 274-289. [Abstract] In psychological studies, it is common to include variables as "control variables," for example, age, gender, educational level, and other similar variables, usually demographic https://blogonresearch.blogspot.com/2008/07/article-potential-problems-in.htmlnoreply@blogger.com (Shu Fai Cheung)0tag:blogger.com,1999:blog-4736329668300304761.post-6398322695987404639Fri, 02 May 2008 15:15:00 +00002010-07-24T11:20:04.602+08:00Factor StructureReliabilityScale Development .Alpha { font-weight: bold; background: #DDDDDD; } .wrappertable { border-width:thin; border-style:solid; border-color:#AAAAAA; padding:0; } .wrappertable td { border-width:thin; border-style:solid; border-color:#AAAAAA; padding:0; } .loading { border-width:thin; border-style:solid; border-color:#AAAAAA; padding:0; } .loading td { border-width:thin; border-style:solid; https://blogonresearch.blogspot.com/2008/05/high-cronbach-alpha-implies-one-single.htmlnoreply@blogger.com (Shu Fai Cheung)0tag:blogger.com,1999:blog-4736329668300304761.post-5176157420932303081Sun, 20 Apr 2008 15:30:00 +00002014-04-12T17:54:58.870+08:00DistributionStatisticsTeachingUpdate 2014-04-12: The techniques below are no longer necessary for Excel. In the recent version of Excel (2010 or later), T.DIST() can return the value of the probability density function. I am not sure about LibreOffice Calc (I now use LibreOffice instead of OpenOffice). A simple (but long) formula I use in Excel and OpenOffice Calc to plot the t-distribution. Recently, for illustration, I https://blogonresearch.blogspot.com/2008/04/plot-t-distribution.htmlnoreply@blogger.com (Shu Fai Cheung)1tag:blogger.com,1999:blog-4736329668300304761.post-5447708968811305180Sat, 12 Apr 2008 17:36:00 +00002014-03-23T01:18:58.269+08:00[I revised the previous article and reposted it on April 2008] The purpose of this blog is to share my articles related to psychology research. I use English as the major language in this blog, as it is the language for most journals in psychology. I originally planned to start a blog mainly for methodology. However, I occasionally have ideas on other aspects of academic research that I would https://blogonresearch.blogspot.com/2008/04/purpose-2008-april.htmlnoreply@blogger.com (Shu Fai Cheung)0tag:blogger.com,1999:blog-4736329668300304761.post-633628494737675537Sun, 06 May 2007 18:19:00 +00002008-07-02T23:18:21.178+08:00Peer ReviewSubmitting ManuscriptsArticle: Tsang, E. W. K., & Frey, B. S. (2007). The as-is journal review process: Let authors own their ideas. Academy of Management Journal, 6, 128-136. [Abstract] In this article, the authors discussed a major problem in the peer-review process: Are the authors really "the authors" of their articles? Are the articles actually the "joint-products" of authors, reviewers, and editors? As the https://blogonresearch.blogspot.com/2007/05/article-as-is-review-by-tsang-and-frey.htmlnoreply@blogger.com (Shu Fai Cheung)0