Last 7 days
Bookmark and Share    
Peer Reviewed
See detailModélisation de la réponse au hasard dans les questionnaires à choix multiples : une approche par modèles de réponses aux items polytomiques
Magis, David ULg

Conference (2017, January 25)

Lorsque l’on doit évaluer les résultats d’un test à l’aide de questions à choix multiples (QCM), le traitement des réponses au hasard (« guessing ») est un enjeu crucial pour une évaluation fiable et non ... [more ▼]

Lorsque l’on doit évaluer les résultats d’un test à l’aide de questions à choix multiples (QCM), le traitement des réponses au hasard (« guessing ») est un enjeu crucial pour une évaluation fiable et non biaisée de la compétence testée. Divers mécanismes (comme l’application d’une correction pour le guessing) ont été proposées mais ne sont pas exemptes de problèmes méthodologiques. Une approche par modèles de réponses aux items (MRI) est aussi souvent considérée, mais revient à résumer les réponses des QCM en « vrai/faux » et appliquer des modèles dichotomiques (comme le modèle logistique à trois paramètres). Le but de cette présentation est double : (a) mettre en évidence les faiblesses d’une telle approche dichotomique, et (b) présenter une modification d’un modèle de réponses aux items polytomiques qui prend en compte le possible « guessing » des participants. Cette approche a l’avantage de modéliser chaque réponse possible (y compris les distracteurs) tout en autorisant une réponse au hasard pour les participants de compétence plus faibles. Une comparaison des approches dichotomique et polytomique est réalisée au moyen de données simulées selon un modèle de réponse au hasard plausible. Les avantages et inconvénients de l’approche polytomique, tant du point de vue technique que pédagogique, sont discutés. [less ▲]

Detailed reference viewed: 17 (0 ULg)
Peer Reviewed
See detailLa détection des réponses inappropriées dans les examens: une étude de 3 indices corrigés selon la méthode de Snijders
Béland, Sébastien; Raîche, Gilles; Magis, David ULg

Conference (2017, January 25)

Certains étudiants peuvent répondre au hasard ou être inattentifs en situation d’examen. Les indices de détection de patrons de réponses inappropriés (person-fit index) sont prometteurs pour détecter ces ... [more ▼]

Certains étudiants peuvent répondre au hasard ou être inattentifs en situation d’examen. Les indices de détection de patrons de réponses inappropriés (person-fit index) sont prometteurs pour détecter ces types de réponse. Nous allons, dans le cadre de cette communication, nous pencher sur le cas de trois indices de détection populaires qui présentent des caractéristiques permettant d’en faciliter l’interprétation : lz, ZU et ZW. Par contre, des études antérieures ont montré que ces trois indices sont fortement affectés par le fait que l’habileté d’un étudiant est estimée plutôt que réelle (Molenaar et Hoijtink, 1990). Snijders (2001) a proposé une version corrigée de l’indice lz (nommée lz*) afin de tenir compte de cette difficulté. De leur côté, Magis, Béland et Raîche (2014) ont corrigé deux autres indices selon l’approche de Snijders (2001) : U* et W*. Il reste cependant à analyser plus en détails le comportement de la famille des indices corrigés, soit, lz*, U* et W*. Pour ce faire, nous effectuons une étude de simulation et une étude des données du test de classement en anglais, langue seconde (TCALS-II). Les résultats démontrent que les indices corrigés lz* et W* sont les plus intéressants à utiliser. De son côté, U* présente des résultats parfois inappropriés, ce qui relance la réflexion sur la pertinence des indices de détection basés sur les carré-moyens non-pondérés. [less ▲]

Detailed reference viewed: 14 (1 ULg)
See detailLes pathologies de l’adaptation et l'adaptation des pathologies
Englebert, Jérôme ULg

Scientific conference (2017, January 21)

Detailed reference viewed: 26 (0 ULg)
Full Text
Peer Reviewed
See detailPhotocleavable stabilizer for the preparation of PHEMA nanogels by dispersion polymerization in supercritical carbon dioxide
Alaimo, David; Grignard, Bruno ULg; Kuppan, Chandrasekar et al

in Polymer Chemistry (2017), 8(3), 581-591

A new photo-sensitive diblock copolymer composed of a hydrophilic sequence of poly(ethylene oxide) (PEO) linked to a CO2-philic sequence of poly(1H,1H,2H,2H-heptadecafluorodecyl acrylate) (PFDA) by a ... [more ▼]

A new photo-sensitive diblock copolymer composed of a hydrophilic sequence of poly(ethylene oxide) (PEO) linked to a CO2-philic sequence of poly(1H,1H,2H,2H-heptadecafluorodecyl acrylate) (PFDA) by a light sensitive o- nitrobenzyl group was successfully synthesized by RAFT polymerization and used as stabilizer for the free radical polymerization of 2-hydroxyethyl methacrylate (HEMA) in dispersion in ?,?,?-trifluorotoluene and supercritical carbon dioxide (scCO2). Thanks to this fluorinated stabilizer, well-defined particles of PHEMA down to 350 nm of diameter were produced in scCO2. Advantageously, the photocleavable group at the block junction of the stabilizer could be cleaved by exposing the particles to UV light so that the fluorinated block could be extracted in TFT or scCO2. As supported by X-ray photoelectron spectroscopy (XPS) analysis, up to 80 % of the fluorinated block of the stabilizer can be removed, leading to efficient swelling and dispersion of the resulting PHEMA nanogels in water. [less ▲]

Detailed reference viewed: 31 (10 ULg)
See detailCharacterization of High Altitude Clouds at the Martian Limb and Terminator Using MAVEN IUVS Observations
Deighan; Stiepen, Arnaud ULg; IUVS team

Conference (2017, January 20)

Detailed reference viewed: 7 (1 ULg)
See detailMars Ozone mapping with MAVEN IUVS
Lefèvre; Stiepen, Arnaud ULg; IUVS team

Conference (2017, January 20)

Detailed reference viewed: 8 (1 ULg)
Full Text
See detailÉtude du cytosquelette de microtubules et implication de la vimentine dans le développement postnatal de la cochlée chez les rongeurs
Renauld, Justine ULg

Doctoral thesis (2017)

According to the World Health Organization, nearly 5% of the population suffers from hearing impairments. These can lead to social isolation, depression, or a reduction in professional abilities. There ... [more ▼]

According to the World Health Organization, nearly 5% of the population suffers from hearing impairments. These can lead to social isolation, depression, or a reduction in professional abilities. There are two types of hearing impairments : conductive hearing impairment and sensorineural hearing impairment. The first results from a defect in sound conduction at the level of the outer or middle ear, the latter results from defects in the cochlea or auditory nerve. In recent years, most studies on the organ of hearing have focused on sensory cells, which are actual receptors of auditory information, whereas supporting cells have been little explored. However, the latter, surrounding the sensory cells, contain a highly developed cytoskeleton, essential to the architecture of the mature organ of Corti. During this thesis work, we first observed the setting up of microtubules within the supporting cells of the organ of Corti. Our morphological data revealed important changes in microtubular cytoskeleton of these cells during the first week of life. We identified an arrangement of microtubules into two distinct bundles. We then demonstrated a modification in the morphology of these microtubules, revealing an increase in the diameter and the number of protofilaments of these cells. We finally demonstrated that tubulin βV has a more specific spatiotemporal distribution than the other tubulins. This characteristic suggests that tubulin βV plays an important role in the maturation of the Corti organ. In the second part of this thesis, we analyzed the involvement of vimentin in cochlear development, and more particularly in the opening of the intercellular spaces and in the myelination of the spiral ganglion, which is the first relay of the auditory pathway. In order to study the role of vimentin in the development of the inner ear, we examined the cochlear morphology and auditory capacities of vimentin deficient mice. Our results indicated that the absence of vimentin does not seem to affect the architecture of the organ of Corti. We then showed a slight morphological difference in the nerve fibers between wildtype and vimentin-knockout mice. We finally checked the hearing physiology of these mice and concluded that vimentin does not play a major role in hearing. [less ▲]

Detailed reference viewed: 12 (4 ULg)
Full Text
See detailUber-like Models for the Electrical Industry
Ernst, Damien ULg

Speech/Talk (2017)

Detailed reference viewed: 225 (16 ULg)
See detailGlobal Simulation of UV atmospheric Emissions
Gonzalez-Galindo; Lopez-Valverde; Forget et al

Poster (2017, January 19)

Detailed reference viewed: 8 (1 ULg)
Full Text
Peer Reviewed
See detailGenetics of Pituitary Tumor Syndromes
Daly, Adrian ULg; BECKERS, Albert ULg

in Melmed, Shlomo (Ed.) The Pituitary (2017)

Detailed reference viewed: 20 (4 ULg)
See detailMoresnet-Neutre (1816-1919): Histoire et mémoire d'une curiosité géopolitique
Brüll, Christoph ULg

Conference given outside the academic context (2017)

Detailed reference viewed: 14 (0 ULg)
See detailA persistent meteoric layer in Mars' atmosphere
Crismani; Schneider; Plane et al

Poster (2017, January 18)

Full Text
See detailFlexible Bayesian Models for Interval-Censored Survival Data
Cetinyürek, Aysun ULg

Doctoral thesis (2017)

Interval-censored time-to-event data arise frequently in clinical trials and longitudinal studies, where the event of interest is only known to have occurred between the two consecutive visits. Interval ... [more ▼]

Interval-censored time-to-event data arise frequently in clinical trials and longitudinal studies, where the event of interest is only known to have occurred between the two consecutive visits. Interval-censoring is a natural generalization of right censored time-to-event data. For right-censored data, extensive number of statistical techniques are available to tackle most research questions under a variety of assumptions. However, for interval-censored data, less well developed procedures are available. A sparse offer in statistical softwares to handle this type of censoring has driven many researchers to use imputation techniques, especially right-point or mid-point imputation. However such imputation strategies can lead to misleading inferences. Our thesis proposes and studies the properties of innovative methods to analyze such data. In the first part of the text, we have extended a Bayesian density estimation procedure for grouped data to estimate hazard ratios and survival functions from interval-censored data. If one further assumes proportionality of the hazards, the proposed strategy also provides estimates of global covariate effects. Clearly, the proposed method provides very good estimates for the regression coefficients and successfully approximates the baseline survival function when the mean interval width is smaller than some threshold defined from the data standard deviation. In a Cox proportional hazards model, the observations are assumed to be independent. However, this may not be true in certain situations where the observed units are clustered or subject to multiple measurements. A number of approaches generalizing Cox's PH model to handle correlated interval-censored data have been proposed in the literature. The shared frailty model is a popular tool to analyze correlated right-censored time-to-event data. Frailty models have also been adapted to handle interval-censored data. In the case of interval-censored time-to-event data, the inclusion of frailties results in complicated intractable likelihoods. In the second part of this thesis, we propose flexible frailty models for analyzing such data by assuming a smooth flexible form for the conditional time-to-event distribution and a parametric or a flexible form for the frailty distribution. It has been indicated in the literature in different contexts that the misspecification of the random effect distribution can influence the estimation of quantities of primary interest, like the fixed effects. To circumvent such misspecification, we have suggested modeling the distribution of the frailty in a flexible way using P-splines or a gamma shape mixture (GSM) distribution. The biggest advantage of using a flexible specification for the density of frailty arises when its shape is of specific interest. If it is considered as a nuisance, assuming a simpler lognormal or gamma frailty would be an adequate solution to draw conclusions related to other model parameters, such as regression coefficients and variance of frailties. Indeed, it was shown in the simulation study that the regression parameter estimates in a shared frailty PH model are robust to the misspecification of the frailty density. Moreover the use of a flexible form for the frailty does not cause any loss of precision in the estimation of regression parameters when compared to the simpler parametric frailty model. Both models provide the possibility to visualize the baseline density and survival functions. Given sufficiently large sample sizes, the flexible approach produces smooth and accurate posterior estimates for the baseline survival function and for the frailty density, and can correctly detect and identify unusual frailty density forms. [less ▲]

Detailed reference viewed: 41 (4 ULg)
Full Text
See detailLeidenfrost effect at its limits
Maquet, Laurent ULg

Doctoral thesis (2017)

Detailed reference viewed: 29 (2 ULg)
Full Text
See detailLes institutions politiques belges
Grandjean, Geoffrey ULg

Learning material (2017)

Detailed reference viewed: 10 (0 ULg)
See detailGlobal UV Imaging by MAVEN/IUVS: Diurnal Cloud Formation, Dust Storms and Atmospheric Scattering.
Schneider; Deighan; Jain et al

Conference (2017, January 17)

Detailed reference viewed: 8 (1 ULg)
See detailThree Types of Aurora observed by MAVEN/IUVS: Implications for Mars’ upper Atmosphere Energy Budget
Connour; Schneider; Jain et al

Poster (2017, January 17)

Detailed reference viewed: 12 (1 ULg)
Full Text
See detailIs Büchi's theorem useful for you? (for an audience of logicians)
Rigo, Michel ULg

Conference (2017, January 17)

Almost a century ago, Presburger showed that the first order theory of the natural numbers with addition is decidable. Following the work of B\"uchi in 1960, this result still holds when adding a function ... [more ▼]

Almost a century ago, Presburger showed that the first order theory of the natural numbers with addition is decidable. Following the work of B\"uchi in 1960, this result still holds when adding a function $V_k$ to the structure, where $V_k(n)$ is the largest power of $k\ge 2$ diving $n$. In particular, this leads to a logical characterization of the $k$-automatic sequences. During the last few years, many applications of this result have been considered in combinatorics on words, mostly by J. Shallit and his coauthors. In this talk, we will present this theorem of B\"uchi where decidability relies on finite automata. Then we will review some results about automatic sequences or morphic words that can be proved automatically (i.e., the proof is carried on by an algorithm). Finally, we will sketch the limitation of this technique. With a single line formula, one can prove automatically that the Thue-Morse word has no overlap but, hopefully, not all the combinatorial properties of morphic words can be derived in this way. We will not assume any background in combinatorics on words from the audience. [less ▲]

Detailed reference viewed: 11 (3 ULg)
Full Text
Peer Reviewed
See detailPresent-day central African forest is a legacy of the 19th century human history
Morin, Julie ULg; Fayolle, Adeline ULg; Favier, Charly et al

in eLife (2017)

The populations of light-demanding trees that dominate the canopy of central African forests are now aging. Here, we show that the lack of regeneration of these populations began ca. 165 ya (around 1850 ... [more ▼]

The populations of light-demanding trees that dominate the canopy of central African forests are now aging. Here, we show that the lack of regeneration of these populations began ca. 165 ya (around 1850) after major anthropogenic disturbances ceased. Since 1885, less itinerancy and disturbance in the forest has occurred because the colonial administrations concentrated people and villages along the primary communication axes. Local populations formerly gardened the forest by creating scattered openings, which were sufficiently large for the establishment of light-demanding trees. Currently, common logging operations do not create suitable openings for the regeneration of these species, whereas deforestation degrades landscapes. Using an interdisciplinary approach, which included paleoecological, archaeological, historical, and dendrological data, we highlight the long-term history of human activities across central African forests and assess the contribution of these activities to present-day forest structure and composition. The conclusions of this sobering analysis present challenges to current silvicultural practices and to those of the future. [less ▲]

Detailed reference viewed: 10 (1 ULg)