PSYCHOMETRIC STUDIO

A Uniform Distribution Made In R Studio

“In probability theory and statistics, the continuous uniform distribution or rectangular distribution is a family of symmetric probability distributions such that for each member of the family, all intervals of the same length on the distribution's support are equally probable.”

SOURCE: https://en.wikipedia.org/wiki/Uniform_distribution_(continuous)

SEE ALSO: http://mathworld.wolfram.com/UniformDistribution.html

http://www.wolframalpha.com/input/?i=uniform+distribution

Here is a UNIFORM DISTRIBUTION I made using R Studio.

Here is my code:

y <-runif(1000,min=-3, max=3)

hist(y,freq=TRUE,main="UNIFORM DISTRIBUTION",breaks=7,col='lightgreen', xlim=c(-3,3), ylim=c(0,350))

Here is a NORMAL DISTRIBUTION I made using R Studio.

Here is my code:

x <-rnorm(1000,mean=0,sd=1)

hist(x,freq=TRUE,main ="NORMAL DISTRIBUTION", breaks=7,col='lightgreen', xlim=c(-3,3), ylim=c(0,350))

Reference: https://www.r-bloggers.com/basics-of-histograms/

Further Reading: http://personality-project.org/r/distributions.html

The Logic Of Using R Project Through SPSS

The logic of using R Project through SPSS is based on the advantages it gives a data analyst in terms of accessing the data set directly and in terms of defining missing values. Here are the quotes from Dalzell (2013) Calling R from SPSS

SOURCE: https://www.ibm.com/developerworks/library/ba-call-r-spss/

“When you use R from SPSS, you can apply R functions to SPSS data while you maintain the integrity of the original database.”

“… although you might want to implement these tools outside of SPSS by exporting the data, data export is never seamless. With the R plug-in you retaining all the features of an SPSS database, particularly the labels of category data and the long descriptors.”

“The great strength of SPSS as a data vault lies in the detailed data dictionary you can create.”

“… SPSS allows greater flexibility in defining missing values than R.”

Convergence In Rasch Software Programs

In Rasch software programs parameter estimates are developed using an iterative procedure which converges on a solution. In my simple way, I describe the procedure as follows: “The item parameter estimates are thrown against the person parameter estimates, and then they are thrown back onto the item parameter estimates, which are them thrown back onto the person parameter estimates etc.”

Linacre (1987) in Rasch Estimation: Iteration and Convergence provides a useful introduction into this topic.

Source: https://www.rasch.org/rmt/rmt11b.htm

He makes a very important point:

“Lack of convergence is an indication that the data do not fit the model well, because there are too many poorly fitting observations. A data set showing lack of convergence can usually be rescued by setting aside for separate study the person or item performances which contain these unexpected responses.”

In my experience, I would also check (re-check) the structure and quality of the dataset in the first instance. This check may find problems with variable coding and data alignment.

Linacre (1987) provides further practical advice in understanding and dealing with convergence and estimation problems.

Advantages Of Using R Project

A precis of Dalzell (2013) Do I need to learn R?

“R is a flexible programming language designed to facilitate exploratory data analysis, classical statistical tests, and high-level graphics. With its rich and ever-expanding library of packages, R is on the leading edge of development in statistics, data analytics, and data mining.”

SOURCE: https://www.ibm.com/developerworks/library/bd-learnr/

Advantages of using R Project include:

• R is a powerful scripting language.
• R leads the way – Many new developments in statistics appear first as R packages before making their way into commercial applications.
• Integration with document publishing – R integrates smoothly with LaTeX document publishing system, meaning that statistical output and graphics from R can be embedded in publication-quality documents.
• No cost

A New Book By Professor Margaret Wu

A new book my Professor Margaret Wu and colleagues is available:

http://www.springer.com/gp/book/9789811033001

Educational Measurement for Applied Researchers: Theory into Practice

Available at:

https://www.amazon.com/Educational-Measurement-Applied-Researchers-Practice/dp/9811033005

See also:

http://www.edmeasurement.com.au/resources

Here are the contents from the chapter on EQUATING:

Plus the chapter on Rasch Measurement Model (Dichotomous Case).

I can't wait to get my copy.

 

Working With The TAM Package Through SPSS v24 Syntax Coding

Working with the TAM Package through SPSS v24 syntax coding.

SOURCE:

https://cran.r-project.org/web/packages/TAM/

http://www.edmeasurement.com.au/resources

http://www.edmeasurementsurveys.com/TAM/Tutorials/

 

STEP 1:

Get R Essentials for SPSS Statistics v24 (aka R Integration Plug in)

https://developer.ibm.com/predictiveanalytics/downloads/

https://github.com/IBMPredictiveAnalytics/R_Essentials_Statistics/releases?cm_mc_uid=70049326057814884026378&cm_mc_sid_50200000=1495098798

(If you have already install R Essentials you may need to uninstall it, as per the support reference below).

Support Reference: http://www-01.ibm.com/support/docview.wss?uid=swg1PI73899

“Steps to perform:

1. Uninstall the R Essentials for Statistics 24 FP1.

2. Download R 3.2.5 from R web site, and install it.

R version 3.2.5 is available from http://www.r-project.org/. You can also download it directly from ftp://ftp.stat.math.ethz.ch/Software/CRAN/bin/windows/base/old/.

3. Reinstall the R Essentials for 24 FP1, and select R 3.2.5 as R home folder upon installation of the Essentials.”

 

STEP 2:

Get R Project version (3.2.5)

The TAM package (2.1-43) needs R Project version 3.2.5.

This older version of R Project can be located here:

https://cran.r-project.org/bin/windows/base/old/

Download the version and install R Project onto your machine.

Check that R Project works.

 

STEP 3:

Get the “TAM” Package (2.1-43) via R Project -

VIA PACKAGES --> INSTALL Package(s)

You also need to load the packages “CDM” and “mvtnorm” -

VIA PACKAGES --> INSTALL Package(s)

Unfortunately, it seems you also need to load the package “polycor”.

https://cran.r-project.org/web/packages/polycor/index.html

So download it as a zip file and then install it -

VIA PACKAGES --> Install package(s) from local zip files…

These 3 packages need to be installed because of the dependencies between them.

Now check that the “TAM” package can be loaded in R Project

VIA PACKAGES --> Load

See below:

Typing the code:

library("TAM")

also achieves the same result.

Loading the “TAM” package seems to work even if loading the “polycor” package by itself causes this statement:

Error: This is R 3.2.5, package ‘polycor’ needs >= 3.3.0

 

STEP 4:

Reinstall R Essentials for Statistics 24 choosing R version 3.2.5.

Then you need to check that the “TAM” package can load in SPSS (this is done via R Essentials).

Call up a new syntax file and write:

BEGIN PROGRAM R.

R.version.string

END PROGRAM.

BEGIN PROGRAM R.

cat("\t\tHello R!\n")

END PROGRAM.

BEGIN PROGRAM R.

#Load TAM library

library("TAM")

END PROGRAM.

Select All and Run Selection (i.e. press the Green Arrow / Green Right Pointing Triangle).

If you have followed these steps, and all is working, you should get this output:

This means you can now start using the “TAM” package by using syntax coding in SPSS v24 (via R Essentials).

 

FURTHER INFORMATION:

https://www.ibm.com/support/knowledgecenter/SSLVMB_24.0.0/spss/programmability_option/r_package_intro.html

What Is Shiny?

What is Shiny?

Shiny uses the R project software environment to create interactive graphics and put them on the web.

See: https://shiny.rstudio.com/

It is based on the R Studio platform:

https://www.rstudio.com/products/rstudio/

An example is here of a useful interactive graphic:

https://gallery.shinyapps.io/CDCPlot/

R Project Tools For Psychometrics

A list of R Project tools for Psychometrics is located here:

https://cran.r-project.org/web/views/Psychometrics.html

(maintained by Dr. Patrick Mair).

Thanks to Dr. Patrick Mair for this really useful resource.

https://psychology.fas.harvard.edu/people/patrick-mair

https://scholar.harvard.edu/mair/publications

I am currently examining the eRm and the TAM packages (see the Item Response Theory - IRT section).

Reproducible Research

Reproducible Research:

"Reproducible research is mainly an organizational principle. Given the original data file and the syntax file, it is possible to re-create every step of the analysis from these two files. Months later, if you need to return to the problem with additional data or a new analysis, it is possible to rebuild the original project. With SPSS, you can maintain a record of every procedure that is run on the data, be it a transformation of the data, the creation of new variables, or an analysis."

Catherine Dalzell (2013) Calling R from SPSS

SOURCE: https://www.ibm.com/developerworks/library/ba-call-r-spss/

Here Dalzell (2013) outlines one of the main advantages of using R packages within the SPSS software environment. Reproducible research is important for scientific progress, secondary data analysis, and professional learning.

See also:

http://reproducibleresearch.net/blog/

http://www.nature.com/authors/policies/availability.html

Updating Your Project Management Knowledge And Skills

A really useful web-site for updating your Project Management knowledge and skills. This site compares the latest trends in Project Management - Agile v Scrum v Waterfall v Kanban.

https://www.smartsheet.com/agile-vs-scrum-vs-waterfall-vs-kanban

I am currently exploring the potential of using Kanban with my Agile skills and processes.

See

https://www.atlassian.com/agile/kanban

https://www.versionone.com/what-is-kanban/

https://leankit.com/learn/kanban/kanban-agile/

http://modusinstitute.com/courses/personal-kanban/lectures/1181169