# Large-Scale Assessments using EdSurvey R Package

AERA 2024 Workshop by AIR

R

NAEP

Tutorial

EdSurvey

## Clustering methods

- jackknife repeated replication (By default TIMSS and NAEP)
- Taylor series approximations
- Hierarchical linear models / Mixture model

## Sampling (takeaway)

- Schools are sampled with probability proportional to size: larger schools are more likely to be selected
- Thus, sample weights should be used to correct the sampling bias
- Student weight is the inverse of the probability of selection

W_{final} = \frac{1}{P_{school}*P_{student}*P_{adj}}

Where

- P_{student} is the probability of one student being selected within one school
- P_{school} is the probability of one school being selected
- P_{adj} are non-participation adjustments

## Plausible Values

- the distribution of latent scores for each individual is estimated by both IRT and latent regression with survey variables
^{1}

^{1} For each student, there are hundreds of context factors, so PCA was used for dimension reduction

## EdSurvey-GPT

EdSurvey-GPT is a chatbot.

### Takeaways:

- Examples of using functions of
`EdSurvey`

- Code debuggging
- Data download using R functions