Publications by Seamus Donnelly
GrammarVocab
1 Preliminaries library(tidyverse) library(brms) library(patchwork) library(rstatix) 1.1 Open datasets and generate summary statistics. These data sets contain the vocabulary and grammar factor scores from the IRT models. vocab <- read_csv("files/vocab.csv") grammar <- read_csv("files/grammar.csv") complete <- left_join(grammar, vocab, by="data_...
1826 sym R (42839 sym/103 pcs) 11 img
Vocab
Note to self: these models take a long time to run, so, rather than estimating them each time I run the code, I’ve saved the models, commented out teh code for running the models, and included the code for opening the models. Be EXTRA CAREFUL that the saved models are the most up to date. 1 Preliminaries library(wordbankr) # WB data library(t...
1571 sym R (54837 sym/77 pcs) 34 img 1 tbl
CDI Complexity 3 LVs
library(wordbankr) # WB data library(tidyverse) # tidy library(mirt) # IRT models library(ltm) # more IRT functions library(psych) # some psychometric stuff (tests of dimensionality) library(Gifi)# some more psychometric stuff (tests of dimensionality) library(knitr) # some formatting, tables, etc library(patchwork) # combining plots. li...
3303 sym R (264459 sym/128 pcs) 40 img 1 tbl
Assignment
1 Some Background Try to use tidyverse style coding for this where possible. library(wordbankr) library(tidyverse) ## ── Attaching packages ─────────────────────────────────────── tidyverse 1.3.0 ── ## ✓ ggplot2 3.3.5 ✓ purrr 0.3.4 ## ✓ tibble 3.0...
3603 sym R (730 sym/8 pcs)
MeetingWithJay
library(wordbankr) # WB data library(tidyverse) # tidy library(mirt) # IRT models library(ltm) # more IRT functions library(psych) # some psychometric stuff (tests of dimensionality) library(Gifi)# some more psychometric stuff (tests of dimensionality) library(knitr) # some formatting, tables, etc library(patchwork) # combining plots. li...
4586 sym R (48254 sym/99 pcs) 30 img 1 tbl
IRT Link Functions
This is a scratch pad. I’m playing around with MIRT to try to identify alternative models to the 2PL. Jay pointed out that the correlation between discrimination and difficulty parameters in the 2PL model may be evidence of a misfitting model. The form of the link function will strongly affect estimation of the extreme values of the latent vari...
6406 sym R (11799 sym/42 pcs) 29 img
Document
The purpose of this analysis is to re-run the DETECT analyses on a set of younger and older participants. If the unidimensional structure of the data in the full dataset is a consequence of mutualism, we would expect this data to intiially be multidimensional and then become unidimensional. options(max.print=500) library(wordbankr) # WB data li...
1255 sym R (45460 sym/301 pcs) 6 img
Document
This script examines whether the vocabulary and grammar sections of the CDI can be distinguished as separate dimensions1. Open the libraries. options(max.print=500) library(wordbankr) # WB data library(tidyverse) # tidy library(mirt) # IRT models library(psych) # some psychometric stuff (tests of dimensionality) library(Gifi)# some more psyc...
1119 sym R (52524 sym/96 pcs) 3 img
Document
This script examines whether the vocabulary and grammar sections of the CDI can be distinguished as separate dimensions. This analysis considers only observations with a unique value on original_id. When we select data for which longitudinal==false there are still many observations with the same values of original_id. Most of these appear to be p...
1676 sym R (38699 sym/113 pcs) 4 img
Document
library(readr) library(knitr) library(tidyverse) library(ggplot2) library(psych) library(sirt) library(Gifi) Inst <- read_rds("Wordbank_Data/Inst_WS_Span.rds") # needs to be saved as rds - csv conflates "" and NA Admin <- read_rds("Wordbank_Data/Admin_WS_Span.rds") N_total = nrow(Admin) N_long = nrow(filter(Admin, longitudinal==TRUE)) ...
179 sym R (5811 sym/32 pcs) 4 img