editor_options | ||||
---|---|---|---|---|
|
- Adds reference
"dutch_gsed2212"
calculated from Dutch data using the "gsed2212" key. Note: This is a temporary name and will be deprecated.
dscore()
anddscore_posterior()
can now copy variables from the inputdata
into the output through theprepend
argument. (#46)- BREAKING CHANGE:
dscore_posterior()
now returns adata.frame
with column names that indicate the quadrature point. This was an unnamedmatrix
. Code that expects amatrix
as the return ofdscore_posterior()
may need to be adapted.
- Per request from CRAN (
Specified C++11: please drop specification unless essential
), removes a C++11 specification
- Sets the default reference in
get_reference()
tophase1
to remain in sync with the defaultkey = "gsed"
- Moves error evasion code into internal
pBCT()
- Document up-rounding to a D-score of 1 or higher when
daz()
andzad()
using the BCT transformation for positive values - Removes the superfluous
names
attribute from the return value ofdaz()
andzad()
- Evades an error produced by internal
pBCT()
whenis.na(nu)
isTRUE
- Renames GSED HH to GSED HF
- Adds instrument
gh1
(GSED-HF, JAN 2023) tobuiltin_itemtable
andbuiltin_itembank
as part of keygsed2212
- Adds example datasets:
sample_sf
,sample_lf
andsample_hf
- Adds vignette to calculate D-scores and DAZ dedicated to GSED instruments
- Renames streams in
gl1
instruments as: aa --> gm, bb --> lg, cc --> fm - Replaces item name
gl1aad001
-->gl1gmd001
, and so on
- Rewrite calls to
select()
andpivot_longer()
to conform totidyselect 1.2.0
grammar
- On 22021130, we found errors in the LF item order. Solves a documentation error. This error was introduced on May 30, 2022 and invalidates keys
gsed2206
andgsed2208
, as well as analyses that rely on correct LF item labels. Version 1.7.0 corrects these problems. - Item labels are taken from
LF1
, corrected using RedCAP comparisons fromPhase_1_master_data_dictionary_V1.0_29_11_2022.xlsx
;LF2
, fromGSED LF Item Guide_October22_FINAL_clean_27Nov22.docx
and manually matched toLF1
.
- Rerun core 293_0 model, check edits, redocument, regenerate diagnostic plots, etc. Check that result is identical.
- Refit full 818_6 model. In general better ICC's, effect on D-score calculation is minor, six items were bad matches
- Introduces new default key
gsed2212
- Introduces new instrument codes
gs1
(GSED SF V1.0) andgl1
(GSED LF V1.0) - Updates
gto
labels with correct order - Set default key to
gsed2212
. This key repairs problems ingsed2206
andgsed2208
. get_labels()
now returns the labels in the same order asitems
- Extends key
gsed2212
with 18 ECDI items using Phase 1 validation data - Updates
builtin_itemtable
andbuiltin_itembank
with correct LF item order - Redocuments upper anchor item
- Solves bug that crashed
dscore::count_mu_phase1(t)
whent
is a vector containingNA
s - Adds example data set
gsample
with 10 cases with SF and LF scores - Adds
order
argument toget_itemnames()
- Repairs an error in the
sort_itemnames()
example - Replaces bitwise by more elegant elementwise comparison in
dscore.cpp
- Removes the dependency on the
sirt
package
- Solves a long-standing issue that led to severe incongruence between LF and SF at the earliest ages (<6M).
- Adds two new keys (
gsed2208
and293_0
) using the Phase 1 validation data for the GSED SF and GSED LF. - Sets
293_0
as the GSED core model and extended it to include the 818 items that fitted the previous model ("gsed2206"). - Adds a new reference (named
phase1
), based on LF and SF data from cohorts GSED-BGD, GSED-PAK and GSED-TZA studies. - **Changes the default key to
gsed2208
and defaultpopulation
to "phase1".**. If you want the old behavior, specify
key = "gsed1912"or
key = "gsed2206"to functions that accept the
keyargument (
dscore(),
dscore_posterior(),
get_age_equivalent(),
get_tau()`).
- Many simplications and update to increase legibility and consistency.
- Makes arguments of
get_age_equivalent()
andget_tau()
consistent withdscore()
- Adapts BCT functions to work with missing and out-of-range data
- Adds support for
phase1
reference toget_reference()
- Adds support for BCT references in
zad()
anddaz()
- Adds
count_mean_phase1()
for setting prior mean equal to the GSED Phase 1 reference - Adds round 2 estimates to
count_mean_phase1()
- Prepares
dscore()
prior_mean functionality to deal with the to-be-implemented new reference as "phase1" - Adds
count_mean_phase1()
function - Adds 18 ECDI items to keys
gsed2206
and294_0
- Initialise proper default
population
when key is294_0
. - Adds experimental key
294_0
to thebuiltin_itembank
. - Adds a
relevance
argument todscore()
anddscore_posterior()
to restrict calculation of D-scores to those items that have theirtau
within this relevance interval around the dynamic EAP estimate. - Repairs bug in
builtin_itembank
that inherits old (gsed1912) tau's into new key (gsed2206) for instrumentsgto
andgpa
. As a consequence, D-score were incorrectly calculated whengto
andgpa
item names were used. - Repairs a bug in
get_age_equivalent()
that resulted in age interval estimates that were too narrow on the D-score scale.
- Adds three new keys (
gsed2206
,lf2206
andsf2206
) that incorporate Phase 1 validation data for the GSED SF, GSED LF and BSID. - Changes the default key to
gsed2206
. If you want the old behavior, specifykey = "gsed1912"
to functions that accept thekey
argument (dscore()
,dscore_posterior()
,get_age_equivalent()
,get_tau()
). - Extends
builtin_itemtable
with names and labels forgpa
(SF),gto
(LF) andgsd
(novel SF) items. - Solves a long-standing issue #29 that gave different D-score under
different transformation. This change affects solutions calculated
with the
metric = "logit"
andtransform
parameters specified. The D-scores calculated under different transforms are now linear transformations of each other.
- Repairs a few problematic outbound URLs
- Updates tau estimates for Mullen items in itembank covering a much wider age range: 65 out of the 85 Mullen items from previous itembank version have new tau estimates; 20 items are removed.
- Updates itembank with new Mullen items. There are 73 new Mullen items added to the new itembank. In total there are now 139 Mullen items in the itembank, estimates are based on PROVIDE and BAMBAM data.
- Moves sources to organisation repository
D-score/dscore
for increased visibility
- Update external links
- Repairs six incorrect BDS numbers in
data-raw/data/bds)edited.csv
- Adds equate groups translated from the
ddata
package tobuiltin_itemtable
- Adds new
itemtable_20200424.txt
with minimal item definitions - Adds 522 missing items that were defined in the
ddata
package - Minimizes the data stored in
builtin_itemtable
- The
get_itemtable()
function gets a new argumentdecompose
for adding info from decomposed item names - Solves warning from
get_itemtable()
whenitemtable
argument is specified
- Uses explicit
drop = TRUE
to account for new behavior intibble 3.0.0
- Adds a facility to create a dynamic (on-the-fly) itemtable from
specified item names in
get_itemtable()
.
- The default rounding is now 2 (D-score), 3 (DAZ). Fixed rounding are now 4 (age) and 4 (p) decimals.
- There was a slight discrepancy in the calculation of DAZ between the
dscore
andclopus
packages. Thedscore
package now first rounds the D-score and then calculate DAZ. - The
dscore()
function now automatically select theprior_mean = ".dutch"
option when the user specifieskey == "dutch"
. This change repair an error in the D-score calculation that occurs when onlykey == "dutch"
was given. - The
dscore()
function now silently handles warnings that may result from taking the log of negative ages.
- Hi, welcome to
dscore 1.0.0
! For the development history, see https://github.com/D-score/dscore - Added a
NEWS.md
file to track changes to the package.