Publications by Jeromy Anglim

Adjusting Correlations for Reliability | Attenuation Formula

28.09.2009

This post discusses ways of adjusting correlations for reliability.Classical Test Theory states that an Observed variable is True Score plus Error. The true score variable is latent. In psychology theoretical interest typically relates more to the latent than the observed variable. How can you estimate the correlation between two lat...

2581 sym 2 img

Blogs on R | Statistics

28.09.2009

What are the current blogs on R? What do they cover?What makes a good blog on R?Of course different readers look for different things, but for me a good blog on R:Provides a Gateway: The blog discovers and shares useful external material and explains why it’s usefulProvides A Forum: The blog raises important issues and topics. Bet...

2485 sym 2 img

Data Mining and Statistics Video Course

03.10.2009

David Mease has an online course presented with complete videos (Statistics 202: Statistical Aspects of Data Mining ). The course uses Excel and R.I might update this post with a few notes below on what is covered as I get a chance to watch the course:The course uses Introduction to Data Mining by Tan, Steinbach, and KumarDay 1(45 minutes): s...

868 sym 2 img

Calculating Scale Scores for Psychological Tests

04.10.2009

This post discusses how to calculate scale scores for multi-item scales. A lot of psychological research uses multi-item scales (e.g., personality tests, symptoms check lists, surveys, etc.). This post focuses on the issues involved with computing these scale scores.This post complements a previous post that looked at computing compos...

6782 sym R (2400 sym/4 pcs) 2 img

Including R Code in a Blog Post

05.10.2009

This post discusses ways of including formatted code in a blog post.I often write posts that include computer code. This might be R code, SPSS syntax, an Inquisit script, or a Visual Basic Macro.The Problem:Initially I did not know how to include this code in a neat form in my posts. The layout was ugly. The font was ugly. And even wo...

1288 sym 2 img

Efficient Variable Selection in R

07.10.2009

This post sets out my procedure for efficiently and reliably selecting variables from a data.frame in RTHE PROBLEMFor years I have been tormented by the hassle of selecting variables in a data.frame. I often work with psychological datasets with hundred or even thousands of variables. Many analyses in psychology will operate on a hund...

3218 sym R (1087 sym/2 pcs) 12 img

Factor Analysis in R

07.10.2009

This post shows an example of running a basic factor analysis in R.Additional Resources:Quick-Rpsych packageJame’s Steiger’s exampleFactoMineR packageThe Example:The example is based on responses by 117 university students to a 50 item version of the IPIP.# Required packages. require(psych); require(foreign); # Import data from S...

841 sym R (24696 sym/2 pcs) 4 img

Comments on "Ecological Statistics with R"

08.10.2009

Didrik Vanhoenacker has put together a site called Ecological Statistics with R.The site provides several resources useful to researchers who are new to R. It gives advice on installing R. It discusses how to use several standard techniques (ANOVA, regression, 2×2 tables, logistic regression, and non-parametric tests). It also cover...

950 sym 2 img

Comments on the "R Clinic"

19.10.2009

Theresa Scott runs an R Clinic at Vanderbilt. Researchers bring their questions along to weekly sessions and some answers are posted on this website.  It’s great to see the combination of statistical consulting services and public posting of answers for the benefit of all.Theresa has also developed a set of course notes on R, R Com...

811 sym 2 img

Data Mining and R

24.10.2009

This post lists a few data mining resources in R. I also provide a few observations on the distinction between data mining, data analysis, and statistics as it pertains to the analysis work that I do in psychology.Online ResourcesThe classic book The Elements of Statistical Learning by Hastie, Tibshirani, Friedman is available for f...

1783 sym 2 img