Psychometric characteristics of cognitive development and achievement instruments in Round 3 of Young Lives

Report

This technical note gives details of the reliability and validity of the tests administered in Round 3 of Young Lives. The tests administered for the Younger Cohort children (aged 7 to 8) were the Peabody Picture of Vocabulary Test (PPVT), Early Grade Reading Assessment (EGRA), and a mathematics test. For the Older Cohort children (aged 14 to 15) the tests administered were the PPVT, sentence comprehension (Cloze), and mathematics. This document gives details of the procedures conducted to clean and build the scores (raw and Rasch) for each of the tests. We give details of the Rasch modelling performed to generate standardised scores and to analyse for possible bias (by gender and language used in the test) in order to estimate and correct for differential item functioning (DIF). The correlation between raw and Rasch scores is estimated, together with demographic variables, in order to check for the contextual validity of these scores for each age group and country. The raw and corrected scores have been uploaded in the Young Lives dataset, which is publicly archived in the UK Public Data Archive. They are also published as a separate Appendix to accompany this document.

Authors: Cueto, S; León, J

• Publication status: Published
• Peer review status: Peer reviewed
• Publisher: Young Lives
• Series: Young Lives Technical Note
• Place of publication: http://www.younglives.org.uk/publications/TN
• Publication date: 2012-01-01
• Edition: Publisher's version
• Paper number: 25
• Language: English
• Keywords: education; methods; child poverty
• Subjects: Early and Child learning; Cognitive development; Children and youth; Education