make_scoring_matrix_aem()has a new argumenteTypeallowing to prepare scoring matrices with distnict processes regarding the extremity trait in different nodes of the decision tree, as desribed in (Merhof & Meiser, 2023).
make_mplus_irtree_model_syntax()andmake_mplus_gpcm_model_syntax()collapses ANALYSIS element to be a single string what makesMplusAutomation::mplusModeler()not failing when run in R >=4.3.0.
make_mplus_irtree_model_syntax()andmake_mplus_gpcm_model_syntax()works withitemsprovided as a simple character vector.
- New functions
make_mplus_irtree_model_syntax()andmake_mplus_gpcm_model_syntax()enabling to conveniently prepare the Mplus syntax specifying response-style models. - New function
read_mplus_object_results()enabling convenient reading results from an estimated mplusObject.
expand_responses()deals with data with responses being a tibble.generate_intercepts_sqn()and thusgenerate_intercepts()acceptsargsdcontaining elements that are not atomic vectors, e.g. list of matrices.
- Considering remarks of Plieninger (2020) regarding the interpretation of IRTree models the distinction between sequential and simultaneous response processes was (almost completely) removed from the package parameter names and descriptions in documentation in favor of labeling them as IRTree and GPCM.
- New functions
make_scoring_matrix_rt()andmake_scoring_matrix_stz()enabling convenient construction of scoring matrices using random thresholds and sum to zero approaches.
- New function
generate_item_expected_scores()that allows to generate (by numerical integration) expected probabilities of responses (response categories) given an item object or test object and covariance matrix of latent traits.- Dependency from the package mvtnorm was added to be able to compute weights for quadrature points.
- New utility functions
thresholds2intercepts()andintercepts2thresholds()allowing conversion between thresholds and intercepts parameterizations of GPCM items.
expand_responses()do not take into account missing values in the data while performing assertion that all values in the data are members of the set of values defined by rownames of thescoringMatrix.generate_intercepts()correctly transforms parameters generated in a parameterization involving item difficulty and thresholds relative to his difficulty into parametrization of intercepts in case of GPCM items (i.e. it cumulatively sums up thresholds and subtracts difficulty instead of adding difficulty to thresholds).- Documentation was corrected to explain, that in case of GPCM mode arguments define how values of thresholds should be generated but that function returns accordingly computed intercepts.
make_item()allows to get non-zero first element of theinterceptsargument if length ofinterceptsequals to the number of rows of thescoringMatrix(previously it returned an error), but still warns that this is atypical specification.
- Test where updated to work correctly after the changes in
generate_intercepts().
make_test()assigns names to the created items by default and provides additionalnamesargument if user wants to provide names himself/herself.generate_test_responses()uses items' names (if there are any) to name columns of the returned matrix.generate_test_responses()converts matrix it returns to numeric one (if only this is possible without loss of information); it also provides additional argumenttryConvertToNumericthat allows to bring back its former behavior (i.e. returning a character matrix).generate_intercepts_sml(), and consequentlygenerate_intercepts()when called withFUNtargument, returns intercepts matrix with additional first columns of zeros to make it compatible with the format that uses functionsimdata()from mirt package (generate_test_responses()was, and still is, able to deal with providing it intercepts either with or without such additional zeros).
- Additional section in README.md describing the way function
simdata()from package mirt may be used to speed up generation of GPCM responses. - Some improvements in documentation.
- Functions that enables user to generate items and tests:
generate_slopes(),generate_intercepts()andmake_test(). - Functions to compute log-normal distribution parameters on the scale of the actual variable (i.e. the exponential scale):
lnorm_mean(),lnorm_sd()andfind_pars_lnorm(). - Function that enables user to generate a very simple scoring matrix that doesn't deal with any response styles:
make_scoring_matrix_trivial(). It is useful if one wants to use rstyles for simulations that don't involve response styles.
- Some improvements in documentation.
- First ready to use version of the package.