• Version
• Download 142
• File Size 162.64 KB
• File Count 1
• Create Date August 2, 2018
• Last Updated August 2, 2018

Assessing literacy on a global scale: Assessment challenges from the Literacy Assessment and Monitoring Programme.

Based on the main survey field observations in Mongolia, it was suspected that some cognitive items show signs of bias for some subgroups, contrary to the expectation that the developed items are neutral. Due to the LAMP two-stage assessment design, some of thetraditional DIF procedures that are commonly used in other large-scale assessment surveys are not suitable here. The Raja’,s DIF index is applied on this dataset. Further research is needed to explore other viable methods for this two-stage test design. Of the 7 Prose and 2 Numeracy items identified in the field observation, not all items suspected to be biased against the focal group are actually statistically biased. This may be explained as the observations in the field arefew (at most 50 cases) while the statistical analysis is based on the general pattern of more than 4000 cases In the next step, it may be of interest to subset the group further to look at how certain items function differently and explore why the differences exist. Coupled with their answer to the background questionnaire, we may understand if this is due to bigger environmental factors like schooling. This exploration may help us understand what extraneousfactors affect literacy skills.

Attached Files