Summer School “Applied Psychometrics in Psychology and Education“

Deadline:  30 May 2019
Open to: participants from around the world
Venue: 05 – 10 August 2019 at HSE University Campus in Moscow, Russia


The participants can opt for one of the following three learning tracks:

Track 1. Rasch Modeling and Behind

The course “Rasch Modeling and Behind” by Prof. Christine Fox and Prof. Svetlana Beltyukova is designed to introduce participants to the principles of objective measurement and build the foundation for using two most common Rasch measurement models: the Rasch model for dichotomous data and the Rasch Rating Scale model. The former is widely used in testing pass/fail situations while the latter is applicable in the analysis of survey or assessment data comprised of ratings. Both models provide rich item and person diagnostics that allow test developers and users to evaluate the extent to which the instrument meaningfully measures intended content in addition to making objective comparisons of scores across different subgroups of interest and over time. Partial Credit Model and Rasch Regression will also be introduced.

Learning Objectives:

  • Verbalize the principles of fundamental measurement
  • Choose and interpret basic Rasch diagnostics
  • Construct and run a WINSTEPS control file for a Dichotomous Model
  • Interpret outputs for a Dichotomous Model
  • Run a WINSTEPS control file for a Rating Scale Model
  • Interpret outputs for a Rating Scale Model
  • Recognize the need for the Partial Credit Model
  • Describe the difference between the Rating Scale and Partial Credit models
  • Choose user-friendly scales and outputs for presenting findings
  • Identify and make decisions about misfitting persons and items
  • Identify and interpret Differential Item Functioning
  • Identify and make decisions about redundant items
  • Construct iterative control file for Rasch Regression
  • Interpret outputs within larger validation framework

Track 1 Schedule

Track 2. Modelling the Growth

This track consists of two courses: “Vertical Scales and Growth Models That Use Them” by Dr. Gary Cook and “Longitudinal Analysis” by Dr. Theodore Walls.

The course “Vertical Scales and Growth Models That Use Them” by Dr. Gary Cook is designed to provide participants with basic concepts and approaches to establishing vertical scales on large scale educational assessments using Item Response Theory (IRT). It also presents methods for applying vertical scales to student growth models. A variety of growth models are presented along with discussions about the benefits and drawbacks of each model.

The course begins with a brief overview of measurement theory starting with Classical Test Theory (CTT) and moving to IRT. The session then discusses methods used to establish vertical scales. For ease of discussion, the one-parameter Rasch models are used, although differences between IRT scaling methodology are mentioned. Following this, an overview of different student growth models using vertical scales is addressed. In this section, theories of action behind each growth model along with methods and resources needed to calculate them are discussed. The course finishes with a discussion about the meaningful application of growth models in educational and research settings.

Learning Objectives:

  • Review basic concepts behind CTT and IRT
  • Understand general approaches to establishing vertical scales using Rasch model horizontal and vertical equating methodologies
  • Understand the benefits and challenges behind different IRT vertical scaling approaches
  • Understand the background and theories of action associated with different educational growth models
  • Understand the basic approaches and methodologies for calculating different educational growth models
  • Have experience calculating different educational growth models using the R statistical software package

Check out more about the Course 

The course “Longitudinal Analysis” by Dr. Theodore Walls covers three important areas in the analysis of longitudinal data: simulation and programming, longitudinal designs, and methods for the analysis of change. The objective of the course is to ensure that graduate students working on research involving longitudinal data have both the baseline conceptual skills and hands-on experience needed to pursue development of applied models in their work. Modeling frameworks include analysis of difference and change scores, repeated measures ANOVA, cross-lagged regression, time series and the family of random coefficient models, including hierarchical linear models, multi-level models, growth curve models, etc. Recently emerging models focused on intensive longitudinal data are also considered.

Learning Objectives:

  • Understand different longitudinal designs
  • Appreciate a range of descriptive and inferential modeling frameworks
  • Experience hands-on modeling toward developing baseline competencies
  • Gain exposure to advanced modeling of intensive longitudinal data, its challenges and opportunities
  • Understand and appreciate literature and community tackling longitudinal data as a field

Check out more about the Course 

A detailed schedule for Track 2 is coming shortly.

Track 3. Simplifying, Identifying, and Evaluating Dimensions: Exploratory and Confirmatory Factor Analyses

The course “Simplifying, Identifying, and Evaluating Dimensions: Exploratory and Confirmatory Factor Analyses” by Dr. Gavin Brown will introduce students to principles and practices behind simplifying data into its component dimensions. This assumes that data contain latent factors and that multiple indicators are used to operationalize those various dimensions. The models rely on stable simultaneous estimation of multiple parameters (incl. variance and covariance matrices of all items, error terms, and relationship structures of latent factors). Students will be taught principles for determining a defensible number of dimensions and how that exploration can be tested. It is a truism that many models of dimensions can fit data; thus, the course will introduce the importance of testing and comparing multiple plausible alternative models. Unsurprising, a small proportion of models will be inadmissible even when N>400. Hence, time will be spent on identifying and troubleshooting problems such as negative error variance and covariance matrix not positive definite. Educational research often seeks to compare participant subgroups who have completed the same test or survey. Nested multigroup invariance testing is a mechanism for determining if scores can be legitimately compared.


Track 1

  • Spoken English
  • Your own laptop

Track 2

  • Spoken English
  • Your own laptop
  • Basic knowledge of measurement concepts such as reliability, validity, generalizability, fairness and equity
  • Basic understanding of CTT and IRT
  • Familiarity with statistics, especially regression models
  • Knowledge of the R statistical software package

Track 3

  • Spoken English
  • Your own laptop
  • Previous exposures to the following statistical methods: chi-square testing; confidence intervals; correlation and covariance; regression

Participation Fee

The School Fee for participants from outside the CIS is RUB 38,000 (approximately EUR 512). Non-CIS enrollees can take advantage of an early bird offer of RUB 30,000 (approximately EUR 405) until April 15, 2019. For applicants who are CIS citizens, a reduced rate of RUB 30,000 is effective up until admissions close on May 30, 2019.

The Fees stated cover tuition for a selected Track, lunch and two coffee breaks daily, and handouts. Travel and accommodation are paid by participants.

Note that you should proceed to pay your School Fee ONLY AFTER a notice is received to confirm that you qualify and have been officially admitted to your chosen School Track (i.e., NOT upon registration).


You have until May 30th, 2019 at 23:59 pm Moscow time to submit your application.

In order to apply, register HERE.

For questions and additional information, please contact the Organizing Committee:

The official web-page.