First, let’s get some wordbank data for American English. We’ll start with production data from the CDI: Words & Gestures (WG) form. We’ll look for DIF based on 1) sex and 2) SES (high vs. low). Questions: How do we choose a dozen anchor items (that we expect to be unbiased)?

action_words animals body_parts clothing descriptive_words food_drink furniture_rooms games_routines household locations outside people pronouns quantifiers question_words sounds time_words toys vehicles
55 36 20 19 37 30 24 19 36 11 27 20 11 8 6 12 8 8 9

We have data from 2164 children for the WG form.

Fit baseline 2PL model

DIF Model: Sex

## Joining, by = c("a1", "definition")
## Joining, by = c("a1", "definition")

Items with a large difference in difficulty across sex

a1 d_m definition d_f d_diff d_diff_abs
3.142711 -17.310308 tomorrow -10.6981329 6.612175 6.612175
3.156998 -17.224483 same -10.7301745 6.494308 6.494308
4.119415 -15.498966 sick -11.7919960 3.706970 3.706970
4.546067 -16.896047 show -13.8793032 3.016743 3.016743
2.700311 -11.209123 morning -8.2999452 2.909178 2.909178
2.937707 -9.633263 dress (object) -6.7470586 2.886205 2.886205
1.899649 -4.685428 doll -2.4533596 2.232068 2.232068
4.856539 -16.536402 hard -14.4381223 2.098280 2.098280
5.702906 -19.077963 living room -16.9897790 2.088184 2.088184
4.823572 -17.788647 today -15.7241449 2.064502 2.064502
2.698688 -11.212253 scared -9.5397046 1.672548 1.672548
1.941153 -8.621930 another -6.9678688 1.654061 1.654061
1.780826 -4.980564 pretty -3.3515742 1.628990 1.628990
4.605111 -13.251055 tired -11.6697246 1.581331 1.581331
5.209194 -16.720621 his -18.2898185 -1.569198 1.569198
2.982405 -7.233030 potty -5.6737735 1.559257 1.559257
3.698646 -10.533709 sleepy -9.0098575 1.523851 1.523851
2.078337 -1.826965 baby -0.3272035 1.499761 1.499761
1.977981 -6.010989 girl -4.5178519 1.493137 1.493137
2.499515 -8.273490 cute -6.7891534 1.484336 1.484336
2.732187 -10.505386 fine -9.0267144 1.478671 1.478671
2.880335 -8.959430 asleep -7.4960716 1.463358 1.463358
3.622350 -11.561431 fast -10.0985227 1.462908 1.462908
3.321156 -12.095334 garden -10.6759239 1.419410 1.419410
2.709747 -6.777470 hug -5.3762452 1.401224 1.401224
2.948435 -8.401435 picture -7.0060893 1.395346 1.395346
3.413015 -5.942904 hair -4.6061459 1.336759 1.336759
3.494207 -9.580445 ride -8.2556266 1.324818 1.324818
2.032135 -5.850055 nice -4.5328440 1.317211 1.317211
5.739061 -19.183899 take -17.8767423 1.307156 1.307156
2.065565 -6.165085 happy -4.8605114 1.304574 1.304574
2.392401 -8.390417 lady -7.1184928 1.271924 1.271924
3.046060 -7.721782 wet -6.4533211 1.268461 1.268461
2.968252 -7.952298 sleep -6.6913232 1.260975 1.260975
2.368755 -6.394353 read -5.1623238 1.232030 1.232030
3.435437 -12.418995 other -11.2022688 1.216727 1.216727
2.943845 -8.957535 watch (action) -7.7409518 1.216584 1.216584

Plot item ease for males vs. females

We label the 37 items with absolute ease difference of more than median + 1SD = 1.21.