The Nominal Data Standard

The data standard

The IRW nominal data standard is meant to describe data wherein responses are nominal and not ordinal. These tables form the nominal data family (source = "nom"). There are two distinct types of such data. The first type involves responses in a necessarily finite (but nominal) set of mutually exclusive categories; a typical example here might be the specific option from a multiple choice item that is selected. The second type involves responses that are potentially infinite; examples here would be constructed responses like essays.

text: This is the nominal version of resp.

Where a nominal table has a counterpart in the main branch of the IRW, text holds the same content that the standard table carries as resp_raw (see the IRW data standard). The two branches differ in what they treat as the response: the standard table scores it and keeps the original alongside, while the nominal table takes the original as the response itself. For the ENEM tables, for instance, the standard table’s resp is 0/1 for incorrect/correct with resp_raw giving the letter the candidate marked, and the nominal table’s text is that letter. Not every nominal table is derived this way — constructed responses such as essays originate in the nominal branch.

Accessing the nominal data

Documentation related to the tables can be found here.

What the family holds

The family holds 140 tables, 2.8 billion responses to 32,074 items. One source dominates: the 52 ENEM tables (Brazil’s national secondary exam, one table per year and test booklet, each holding the letter every candidate marked) account for 83% of the responses. The remaining tables are smaller and more varied: option-level records of multiple-choice tests, stimulus-identification tasks, and 12 tables of constructed responses (essays, free-text answers, drawn responses) where the number of distinct categories runs into the hundreds or thousands because nearly every response is unique. Each table’s own counts are in irw_metadata(source = "nom").

Figure 1: Nominal tables by number of distinct response categories. The 20+ group is the constructed-response tables.

Every table in the family

Each table links to its own page, with citation, licence and fetch snippets.

Table	Respondents	Items	Responses	Categories	Licence	Source
`asap20train_nom`	17,307	7	17,307	17,307	CC BY 4.0	link
`blum_2018_imak_nom`	317	23	7,291	10	CC BY 4.0	link
`borges_brazil_residency_2024_cbt_nom`	3,223	120	386,760	5	CC0 1.0	link
`borges_brazil_residency_2024_pbt_nom`	1,994	100	199,400	6	CC0 1.0	link
`choi_2020_ednet_listening_nom`	637,446	4,663	40,745,825	4	CC BY-NC 4.0	link
`choi_2020_ednet_reading_nom`	723,232	7,620	54,520,455	4	CC BY-NC 4.0	link
`cifar10h_nom`	2,571	10,000	514,200	10	CC BY-NC-SA 4.0	link
`cos101_2026_nom`	294	3	882	873	CC BY 4.0	link
`encceja_2018_ef_ci_nom`	115,838	30	3,475,140	6	CC BY 4.0	link
`encceja_2018_ef_hg_nom`	114,673	30	3,440,190	6	CC BY 4.0	link
`encceja_2018_ef_lp_nom`	121,871	30	3,656,130	6	CC BY 4.0	link
`encceja_2018_ef_ma_nom`	118,490	30	3,554,700	6	CC BY 4.0	link
`encceja_2018_em_ch_nom`	529,921	30	15,897,630	6	CC BY 4.0	link
`encceja_2018_em_cn_nom`	539,515	30	16,185,450	6	CC BY 4.0	link
`encceja_2018_em_lc_nom`	580,975	30	17,429,250	6	CC BY 4.0	link
`encceja_2018_em_mt_nom`	574,651	30	17,239,530	6	CC BY 4.0	link
`encceja_2019_ef_ci_nom`	167,684	30	5,030,520	6	CC BY 4.0	link
`encceja_2019_ef_hg_nom`	166,588	30	4,997,640	6	CC BY 4.0	link
`encceja_2019_ef_lp_nom`	175,184	30	5,255,520	6	CC BY 4.0	link
`encceja_2019_ef_ma_nom`	169,177	30	5,075,310	6	CC BY 4.0	link
`encceja_2019_em_ch_nom`	795,299	30	23,858,970	6	CC BY 4.0	link
`encceja_2019_em_cn_nom`	787,292	30	23,618,760	6	CC BY 4.0	link
`encceja_2019_em_lc_nom`	817,967	30	24,539,010	6	CC BY 4.0	link
`encceja_2019_em_mt_nom`	874,787	30	26,243,610	6	CC BY 4.0	link
`encceja_2020_ef_ci_nom`	37,216	30	1,116,480	6	CC BY 4.0	link
`encceja_2020_ef_hg_nom`	38,392	30	1,151,760	6	CC BY 4.0	link
`encceja_2020_ef_lp_nom`	39,483	30	1,184,490	6	CC BY 4.0	link
`encceja_2020_ef_ma_nom`	41,606	30	1,247,909	6	CC BY 4.0	link
`encceja_2020_em_ch_nom`	294,380	30	8,831,400	6	CC BY 4.0	link
`encceja_2020_em_cn_nom`	278,970	30	8,368,180	6	CC BY 4.0	link
`encceja_2020_em_lc_nom`	306,079	30	9,182,370	6	CC BY 4.0	link
`encceja_2020_em_mt_nom`	377,729	30	11,328,051	6	CC BY 4.0	link
`encceja_2022_ef_ci_nom`	52,436	30	1,573,080	6	CC BY 4.0	link
`encceja_2022_ef_hg_nom`	53,949	30	1,618,470	6	CC BY 4.0	link
`encceja_2022_ef_lp_nom`	55,899	29	1,621,071	6	CC BY 4.0	link
`encceja_2022_ef_ma_nom`	54,837	29	1,590,273	6	CC BY 4.0	link
`encceja_2022_em_ch_nom`	380,869	30	11,426,070	6	CC BY 4.0	link
`encceja_2022_em_cn_nom`	361,759	30	10,850,270	6	CC BY 4.0	link
`encceja_2022_em_lc_nom`	391,408	30	11,742,240	6	CC BY 4.0	link
`encceja_2022_em_mt_nom`	419,625	30	12,588,750	6	CC BY 4.0	link
`encceja_2023_ef_ci_nom`	33,450	30	1,003,500	6	CC BY 4.0	link
`encceja_2023_ef_hg_nom`	34,326	30	1,029,780	6	CC BY 4.0	link
`encceja_2023_ef_lp_nom`	37,583	30	1,127,490	6	CC BY 4.0	link
`encceja_2023_ef_ma_nom`	35,869	30	1,075,476	6	CC BY 4.0	link
`encceja_2023_em_ch_nom`	237,696	30	7,130,880	6	CC BY 4.0	link
`encceja_2023_em_cn_nom`	236,773	30	7,103,190	6	CC BY 4.0	link
`encceja_2023_em_lc_nom`	262,973	30	7,889,190	6	CC BY 4.0	link
`encceja_2023_em_mt_nom`	299,267	30	8,976,146	6	CC BY 4.0	link
`encceja_2024_ef_ci_nom`	20,396	30	611,880	6	CC BY 4.0	link
`encceja_2024_ef_hg_nom`	21,200	30	636,000	6	CC BY 4.0	link
`encceja_2024_ef_lp_nom`	23,462	30	703,860	6	CC BY 4.0	link
`encceja_2024_ef_ma_nom`	21,993	30	659,600	6	CC BY 4.0	link
`encceja_2024_em_ch_nom`	166,794	30	5,003,820	6	CC BY 4.0	link
`encceja_2024_em_cn_nom`	162,446	30	4,873,380	6	CC BY 4.0	link
`encceja_2024_em_lc_nom`	181,453	30	5,443,590	6	CC BY 4.0	link
`encceja_2024_em_mt_nom`	220,567	30	6,616,070	6	CC BY 4.0	link
`encceja_2025_ef_ci_nom`	17,997	30	539,910	6	CC BY 4.0	link
`encceja_2025_ef_hg_nom`	18,237	29	528,873	6	CC BY 4.0	link
`encceja_2025_ef_lp_nom`	19,581	30	587,413	6	CC BY 4.0	link
`encceja_2025_ef_ma_nom`	19,104	28	534,897	6	CC BY 4.0	link
`encceja_2025_em_ch_nom`	151,857	30	4,555,710	6	CC BY 4.0	link
`encceja_2025_em_cn_nom`	149,336	30	4,480,080	6	CC BY 4.0	link
`encceja_2025_em_lc_nom`	161,847	29	4,693,563	6	CC BY 4.0	link
`encceja_2025_em_mt_nom`	186,322	29	5,403,194	6	CC BY 4.0	link
`enem_2013_1mil_ch_nom`	1,000,000	45	45,000,000	7	CC BY 4.0	link
`enem_2013_1mil_cn_nom`	1,000,000	45	45,000,000	7	CC BY 4.0	link
`enem_2013_1mil_lc_nom`	1,000,000	50	45,000,000	7	CC BY 4.0	link
`enem_2013_1mil_mt_nom`	1,000,000	45	45,000,000	7	CC BY 4.0	link
`enem_2014_1mil_ch_nom`	1,000,000	45	45,000,000	7	CC BY 4.0	link
`enem_2014_1mil_cn_nom`	1,000,000	45	45,000,000	7	CC BY 4.0	link
`enem_2014_1mil_lc_nom`	999,843	50	44,992,935	7	CC BY 4.0	link
`enem_2014_1mil_mt_nom`	999,844	45	44,992,980	7	CC BY 4.0	link
`enem_2015_1mil_ch_nom`	1,000,000	45	45,000,000	7	CC BY 4.0	link
`enem_2015_1mil_cn_nom`	1,000,000	45	45,000,000	7	CC BY 4.0	link
`enem_2015_1mil_lc_nom`	999,906	50	44,995,770	7	CC BY 4.0	link
`enem_2015_1mil_mt_nom`	999,907	45	44,995,815	7	CC BY 4.0	link
`enem_2016_1mil_ch_nom`	1,000,000	45	45,000,000	7	CC BY 4.0	link
`enem_2016_1mil_cn_nom`	1,000,000	45	44,997,972	7	CC BY 4.0	link
`enem_2016_1mil_lc_nom`	1,000,000	50	45,000,000	7	CC BY 4.0	link
`enem_2016_1mil_mt_nom`	1,000,000	45	44,999,610	7	CC BY 4.0	link
`enem_2017_1mil_ch_nom`	1,000,000	45	44,978,352	7	CC BY 4.0	link
`enem_2017_1mil_cn_nom`	999,471	45	44,975,131	7	CC BY 4.0	link
`enem_2017_1mil_lc_nom`	1,000,000	50	44,979,936	7	CC BY 4.0	link
`enem_2017_1mil_mt_nom`	999,471	45	44,976,062	7	CC BY 4.0	link
`enem_2018_1mil_ch_nom`	1,000,000	45	44,999,858	7	CC BY 4.0	link
`enem_2018_1mil_cn_nom`	1,000,000	45	44,998,580	7	CC BY 4.0	link
`enem_2018_1mil_lc_nom`	1,000,000	50	45,000,000	7	CC BY 4.0	link
`enem_2018_1mil_mt_nom`	1,000,000	44	43,999,330	7	CC BY 4.0	link
`enem_2019_1mil_ch_nom`	1,000,000	45	45,000,000	7	CC BY 4.0	link
`enem_2019_1mil_cn_nom`	1,000,000	45	44,999,716	7	CC BY 4.0	link
`enem_2019_1mil_lc_nom`	1,000,000	50	45,000,000	7	CC BY 4.0	link
`enem_2019_1mil_mt_nom`	1,000,000	45	44,999,858	7	CC BY 4.0	link
`enem_2020_1mil_ch_nom`	1,000,000	45	45,000,000	7	CC BY 4.0	link
`enem_2020_1mil_cn_nom`	998,883	44	43,949,264	7	CC BY 4.0	link
`enem_2020_1mil_lc_nom`	1,000,000	50	45,000,000	7	CC BY 4.0	link
`enem_2020_1mil_mt_nom`	998,883	44	43,947,884	7	CC BY 4.0	link
`enem_2021_1mil_ch_nom`	1,000,000	45	45,000,000	7	CC BY 4.0	link
`enem_2021_1mil_cn_nom`	999,846	45	44,991,368	7	CC BY 4.0	link
`enem_2021_1mil_lc_nom`	1,000,000	50	44,997,422	7	CC BY 4.0	link
`enem_2021_1mil_mt_nom`	999,846	44	43,992,795	7	CC BY 4.0	link
`enem_2022_1mil_ch_nom`	1,000,000	45	45,000,000	7	CC BY 4.0	link
`enem_2022_1mil_cn_nom`	999,438	45	44,974,302	7	CC BY 4.0	link
`enem_2022_1mil_lc_nom`	1,000,000	50	44,998,364	7	CC BY 4.0	link
`enem_2022_1mil_mt_nom`	999,438	44	43,974,507	7	CC BY 4.0	link
`enem_2023_1mil_ch_nom`	1,000,000	45	45,000,000	7	CC BY 4.0	link
`enem_2023_1mil_cn_nom`	999,722	45	44,986,496	7	CC BY 4.0	link
`enem_2023_1mil_lc_nom`	1,000,000	50	44,999,136	7	CC BY 4.0	link
`enem_2023_1mil_mt_nom`	999,722	44	43,986,774	7	CC BY 4.0	link
`enem_2024_1mil_ch_nom`	1,000,000	45	45,000,000	7	CC BY 4.0	link
`enem_2024_1mil_cn_nom`	999,917	44	43,995,909	7	CC BY 4.0	link
`enem_2024_1mil_lc_nom`	1,000,000	50	45,000,000	7	CC BY 4.0	link
`enem_2024_1mil_mt_nom`	999,917	45	44,994,948	7	CC BY 4.0	link
`enem_2025_1mil_ch_nom`	980,930	45	44,141,850	7	CC BY 4.0	link
`enem_2025_1mil_cn_nom`	980,793	43	42,168,273	7	CC BY 4.0	link
`enem_2025_1mil_lc_nom`	980,930	50	44,141,850	7	CC BY 4.0	link
`enem_2025_1mil_mt_nom`	980,793	44	43,143,240	7	CC BY 4.0	link
`experimental_iq_nom`	399	25	9,474	8	CC BY 4.0	link
`fitz_2024_numeracy_nom`	2,806	11	30,722	1,315	CC BY 4.0	link
`geography_nom`	91,331	1,458	10,087,305	1,455	ODbL 1.0	link
`goldberg_2018_spa_computer_use_nom`	709	2	1,388	12	CC0 1.0	link
`hachenberger_2025_stroop_main_nom`	70	2	236,196	4	CC BY 4.0	link
`hachenberger_2025_stroop_pilot_nom`	21	2	47,695	4	CC BY 4.0	link
`himmelstein-berlin_numeracy-2025_nom`	1,147	4	4,588	4	CC BY 4.0	link
`mthimkhulu_2023_pirls_reading_nom`	1,680	4	5,486	4	CC BY 4.0	link
`much_tte_2025_concentrationtask_nom`	1,244	9	11,190	16	CC BY 4.0	link
`much_tte_2025_matrixreasoning_nom`	1,244	20	21,761	1,004	CC BY 4.0	link
`myszkowski_2018_spmls_nom`	499	12	5,988	8	CC BY 4.0	link
`nomt_hooper_2024_study2_nom`	393	144	56,592	3	CC BY 4.0	link
`papousek_2017_anatomy_nom`	18,563	2,894	1,182,065	2,898	ODbL 1.0	link
`persuade_learningagency_nom`	15,594	15	15,594	15,594	CC BY 4.0	link
`preference_inventory_nom`	13,502	12	162,024	4	CC BY 4.0	link
`psychtools_ability_nom`	1,509	16	23,257	8	GPL-3.0	link
`semeval2013_scientsbank_nom`	3,026	195	10,768	5	CC BY-SA 3.0	link
`sirota_2018_crt_nom`	452	7	3,164	3	CC BY 4.0	link
`suarez_2026_statistics_nom`	431	85	6,658	306	CC BY 4.0	link
`vocabulary_iq_nom`	12,173	75	543,002	668	CC BY 4.0	link
`voropaeva_2026_health_media_nom`	121	21	2,298	53	CC BY 4.0	link
`wilmer-mrmet-normative-data-set-2022_nom`	9,295	37	343,876	40	CC BY 4.0	link
`wilmer-rmet-normative-data-set-2022_nom`	17,680	36	636,480	77	CC BY 4.0	link
`zorowitz_2023_marsib_nom`	1,501	384	23,596	4	MIT	link

Showing 140 of 140 tables. Fetch any of them with irw_fetch("<table>", source = "nom") or irw.fetch("<table>", source="nom").

An example

Let’s look at an example of how we can fit the nominal response model to a table (preference_inventory_nom) from the nominal branch of the IRW and generated category response functions for all items.

Code

df<-irw::irw_fetch("preference_inventory_nom",source='nom')
resp<-irw::irw_long2resp(df,resp_col='text')
resp$id<-NULL
library(mirt)
m<-mirt(resp,1,'nominal')
f<-function(mod,n) {
    extr <- extract.item(mod,n)
    Theta <- matrix(seq(-6,6, length.out=2000))
    pr <- probtrace(extr, Theta)
    list(Theta,pr)
}
out<-list()
for (n in 1:ncol(resp)) out[[n]]<-f(m,n)
##
par(mfrow=c(4,3),mgp=c(2,1,0),mar=c(3,3,1,1))
cols<-c("black","red","blue","green")
for (i in 1:length(out)) {
    plot(NULL,xlim=c(-5,5),ylim=0:1,xlab='theta',ylab='Pr')
    z<-out[[i]]
    for (j in 1:4) lines(z[[1]],z[[2]][,j],col=cols[j])
}