RALSA {RALSA}R Documentation

R Analyzer for Large-Scale Assessments (RALSA)

Description

The RALSA package provides functionality for analyzing data from large-scale assessments and surveys which use complex sampling and assessment design. Such (international) assessments and surveys are TIMSS, PIRLS and PISA, for example.

The sampling (complex sampling design) in large-scale assessments and surveys is multistage with probability proportional to the size of the (primary) sampling units (usually schools), i.e. with unequal probabilities of selection. Thus, all weights assigned to the individual respondents reflect these unequal probabilities. This is quite different from the usual simple or systematic random sampling. Different modifications of Jackknife Repeated Replication (JRR, with full or half replication) or Balanced Repeated Replication (BRR) are used in different studies to compute the standard errors of the population estimates. The proficiency test scores (complex assessment design) is applied to cope with practical issues. No respondent takes all test items, but the items are distributed across multiple test item blocks and the blocks are rotated across multiple assessment booklets, each respondent taking one booklet only. As a consequence, no respondent receives a single test score, but five (or even 10) separate test scores (called "plausible values" or PVs) resulting from multiple imputation technique where the missing by design responses are imputed. As a consequence of the complex sampling and assessment designs, each estimate has to be computed with each JRR or BRR weight and each PV (this can take up to 781 computations per estimate per group per country, depending on the study), then summarized to compute the final estimate, its sampling and imputation variance, and the final standard error.

RALSA provides data preparation and analysis functions which take into account the complex sampling and assessment design of the studies. Each study has its a different implementation of the complex sampling and assessment designs and RALSA handles these and implements the corresponding computational procedure.

Studies

Currently, RALSA works with data from all cycles of the the following studies:

More studies (national international) will be added in future.

Functions

Currently, RALSA provides the following functionality:

The lsa.pcts.means, lsa.prctls, lsa.bench and lsa.crosstabs also have the option to produce graphs from the estimates.

More studies and analysis types will be added in future, and the existing ones will be updated, adding more features.

RALSA also has a Graphical User Interface (GUI) for the less technical users. The GUI incorporates all aspects of the data preparation and analysis functions.

Author(s)

Plamen V. Mirazchiyski, INERI

References

Here are the two articles presenting the package and it's technical details:

Mirazchiyski, P.V. (2021). RALSA: The R analyzer for large-scale assessments. Large-scale Assess Educ 9(21), 1-24. https://doi.org/10.1186/s40536-021-00114-4

Mirazchiyski, P. V. (2021). RALSA: Design and Implementation. Psych, 3(2), 233-248. https://doi.org/10.3390/psych3020018

Here is a list of selected references related to some of the studies' design, relevant to their latest cycles:

Foy, P., & LaRoche, S. (2017). Estimating Standard Errors in the PIRLS 2016 Results. In M. O. Martin, I. V. S. Mullis, & M. Hooper (Eds.), Methods and Procedures in PIRLS 2016 (p. 4.1-4.22). Lynch School of Education, Boston College.

Foy, P., & Yin, L. (2016). TIMSS 2015 Achievement Scaling Methodology. In M. O. Martin, I. V. S. Mullis, & M. Hooper (Eds.), Methods and Procedures in TIMSS 2015 (p. 13.1-13.62). TIMSS & PIRLS International Study Center.

LaRoche, S., Joncas, M., & Foy, P. (2016). Sample Design in TIMSS 2015. In M. O. Martin, I. V. S. Mullis, & M. Hooper (Eds.), Methods and Procedures in TIMSS 2015 (p. 3.1-3.37). TIMSS & PIRLS International Study Center.

OECD. (in press). PISA 2018 Technical Report. OECD.

Rutkowski, L., Gonzalez, E., Joncas, M., & von Davier, M. (2010). International Large-Scale Assessment Data: Issues in Secondary Analysis and Reporting. Educational Researcher, 39(2), 142-151.

Rutkowski, L., Rutkowski, D., & von Davier, M. (2014). A Brief Introduction to Modern International Large-Scale Assessment. In L. Rutkowski, M. von Davier, & D. Rutkowski (Eds.), Handbook of International Large-Scale Assessments: Background, Technical Issues, and Methods of Data Analysis (pp. 3-10). CRC Press.


[Package RALSA version 1.4.7 Index]