Skip to contents

Numeracy test data for the DIF and fairness lab

Source: R/labs.R
numeracy.Rd

Simulated right/wrong responses of 2,000 Grade 6 students, 500 of them English language learners (ELL), to a fictional 30-item provincial numeracy test. Some items are word problems, some are bare calculations. DIF was planted on purpose for the fairness lab; lab_truth("numeracy") lists it.

Usage

numeracy

numeracy_items

Format

numeracy is a data frame with 2,000 rows and 33 columns: id (student ID), ell (factor, Fluent or ELL), gender (factor, Girl or Boy) and N01 to N30 (1 = right, 0 = wrong). numeracy_items describes the items: item, curriculum strand, format (Symbolic or Word problem) and words (word count).

An object of class data.frame with 30 rows and 4 columns.

Source

Simulated; the script is data-raw/numeracy.R in the package's GitHub repository.

Examples

head(numeracy[, 1:6])
#>      id    ell gender N01 N02 N03
#> 1 S0001 Fluent    Boy   0   1   1
#> 2 S0002 Fluent    Boy   0   0   1
#> 3 S0003 Fluent   Girl   0   0   0
#> 4 S0004 Fluent   Girl   0   1   1
#> 5 S0005 Fluent    Boy   0   1   1
#> 6 S0006    ELL    Boy   0   0   1
tapply(rowSums(numeracy[, -(1:3)]), numeracy$ell, mean)
#> Fluent    ELL 
#> 14.734 11.264