UNI_ASK_U
In non-cognitive measurement, certain traits such as clinical or psychiatric symptoms, addictions, extreme beliefs or maladaptive personality
traits can be plausibly modeled as unipolar dimensions (Lucke, 2014, 2015; Reise & Waller, 2009). The low end of the dimension merely reflects
the absence of trait manifestations while the high end reflects different levels of trait severity or extremeness. Furthermore, it is reasonable
to scale this type of dimensions as adopting only positive values and to assume that they are more informative and meaningful at the upper end.
When traits of this type are measured in community populations, an expected result is that the items and test scores have extreme (rightly skewed)
distributions, because most individuals are expected to have low trait levels and be piled-up at the lower end.
UNI_ASK_U is designed to fit unidimensional unipolar Item Response Theory (IRT) models for continuous response items in which the latent trait
distribution is assumed to be lognormal (0,1). The program implements two models based on a power-type link function (Lucke, 2015): (a) the Log-Logistic
Continuous Response Model (LL-CRM; Ferrando et al., 2024), and (b) the Weibull Continuous Response Model (Wb-CRM; Ferrando et al., 2025). Both models can
be viewed as transformed versions of standard continuous-response IRT models, specifically adapted for unipolar constructs with asymmetric item response
functions and skewed latent distributions.
The LL-CRM uses a log-logistic link function, while the Wb-CRM employs a Weibull cumulative distribution function (complementary log-log link when linearized)
as the item response function. Both models feature a two-stage estimation procedure: (1) item calibration via factor analysis of the transformed scores, and (2)
individual scoring using Expected A Posteriori (EAP) estimation with lognormal priors. The program provides a comprehensive output including item parameters
(easiness, discrimination, difficulty), multidimensional slopes and locations, individual trait estimates with posterior standard deviations, and reliability indices.
UNI_ASK_U has been developed in R version 4.5.1 and runs with R versions more recent than 3.5.0. It operates in any system that supports R (Windows, Linux, Mac OS).
The package includes both a console-based script (UNI_ASK_U.R) and a Shiny-based graphical user interface (UNI_ASK_U_GUI.R) for easier interaction. If you think the
program can be useful in your research or practice, please give it a try. You can freely download the package and a detailed manual here. Working examples are also
included to illustrate the program's functionality.
References
Ferrando, P. J., Morales-Vives, F., & Hernandez-Dorado, A. (2024). Measuring unipolar traits with continuous-response items: Some methodological and substantive developments. Educational and Psychological Measurement, 84(3), 425-449. https://doi.org/10.1177/00131644231181889
Ferrando, P. J., Morales-Vives, F., Casas, J. M., & Navarro-González, D. (2025). A Weibull-Link Response Model for measuring Unipolar-Skewed Constructs with Continuous Responses. [Journal details pending]
Lucke, J. F. (2014). Positive trait item response models. In R. E. Millsap, L. A. van der Ark, D. M. Bolt, and C. M. Woods (Eds.), New developments in quantitative psychology (pp. 199–213). Springer.
Lucke, J. F. (2015). Unipolar item response models. In S. P. Reise and D. A. Revicki (Eds.), Handbook of item response theory modeling: Applications to typical performance assessment (pp. 272–284). Routledge/Taylor & Francis Group. https://doi.org/10.4324/9781315736013Reise, S. P., & Waller, N. G. (2009). Item response theory and clinical measurement. Annual Review of Clinical Psychology, 5, 27-48. https://doi.org/10.1146/annurev.clinpsy.032408.153553