A-61 The Cattell-Horn-Carroll Model Does Not Outperform a Traditional Neuropsychological Model When Using a Digital Test Battery

Objective Following recent (meta-)analyses showing that the Cattell-Horn-Carroll (CHC) model outperforms traditional neuropsychological models of cognition, we investigated the relative performance of the traditional neuropsychological model (TNM) of cognitive functioning and the CHC model using the digital Philips IntelliSpace Cognition (ISC) test battery. Method The ISC battery consists of 11 conventional cognitive tests, which were administered to 221 healthy participants. We used confirmatory factor analysis (CFA) to map outcome measures to cognitive domains. Both models contained processing speed, working memory, memory, and visuospatial ability. TNM included executive functioning consisting of TMT B, Stroop interference, and Phonetic and Semantic Fluency. In the CHC model, these outcome measures were distributed over processing speed and a word fluency domain. The models’ fit was compared through Vuong test for non-nested models. Results Both models were a good fit (TNM: χ2(64) = 91.95, p = .013, CFI = .971, RMSEA = .044; CHC: χ2(64) = 83.66, p = .050, CFI = .979, RMSEA = .037). The Vuong test showed that the models could be significantly distinguished based on the observed data, ω2 = 0.21, p < .001. However, the non-nested likelihood ratio test did not offer evidence that either model was a better fit, z = -0.61, p = .727. The 95% confidence interval of the AIC difference contained zero [-18.60, 35.20]. Conclusions Recent findings showing that the CHC model outperforms the TNM are not corroborated by our findings. Instead, both models captured the data equally well, suggesting that the two theoretical frameworks are not necessarily mutually exclusive.

A test of inflated zeros for Poisson regression models

Excessive zeros are common in practice and may cause overdispersion and invalidate inference when fitting Poisson regression models. There is a large body of literature on zero-inflated Poisson models. However, methods for testing whether there are excessive zeros are less well developed. The Vuong test comparing a Poisson and a zero-inflated Poisson model is commonly applied in practice. However, the type I error of the test often deviates seriously from the nominal level, rendering serious doubts on the validity of the test in such applications. In this paper, we develop a new approach for testing inflated zeros under the Poisson model. Unlike the Vuong test for inflated zeros, our method does not require a zero-inflated Poisson model to perform the test. Simulation studies show that when compared with the Vuong test our approach not only better at controlling type I error rate, but also yield more power.

An assessment of resource availability for problem based learning in a Ghanaian University setting

Purpose The purpose of this paper is to assess the differences pertaining to the resources presently accessible for problem-based learning (PBL) among six colleges of Kwame Nkrumah University of Science and Technology (KNUST) in Ghana. Design/methodology/approach Data for the study are the cross-sectional type drawn from 1,020 students. Poisson and zero-inflated Poisson (ZIP) models were implemented on the data to ascertain the variations regarding the extent of resources available for PBL across the colleges of the university. Findings The study outlines the specific resources accessible for PBL across college levels of KNUST. On aggregate, 25.7 per cent reported that their respective colleges have sufficient resources, while 74.3 per cent indicated otherwise. The ZIP model exhibited superiority over the Poisson model, when compared under a Vuong test. As per the ZIP model, none of the colleges appeared to differ significantly in terms of having sufficient resource for PBL. Practical implications Findings are applicable to informed decision-making which targets achieving quality education through the use of PBL. Access to sufficient resources that meet the needs of colleges or departments of a University is emphasized. Originality/value The application of Poisson and ZIP models to aggregated count data in a PBL setting is novel.

A Simple Distribution-Free Test for Nonnested Model Selection

This paper considers a simple distribution-free test for nonnested model selection. The new test is shown to be asymptotically more efficient than the well-known Vuong test when the distribution of individual log-likelihood ratios is highly peaked. Monte Carlo results demonstrate that for many applied research situations, this distribution is indeed highly peaked. The simulation further demonstrates that the proposed test has greater power than the Vuong test under these conditions. The substantive application addresses the effect of domestic political institutions on foreign policy decision making. Do domestic institutions have effects because they hold political leaders accountable, or do they simply promote political norms that shape elite bargaining behavior? The results indicate that the latter model has greater explanatory power.