Classification Performance of Answer-Copying Indices Under Different Types of IRT Models.
ABSTRACT: Test fraud has recently received increased attention in the field of educational testing, and the use of comprehensive integrity analysis after test administration is recommended for investigating different types of potential test frauds. One type of test fraud involves answer copying between two examinees, and numerous statistical methods have been proposed in the literature to screen and identify unusual response similarity or irregular response patterns on multiple-choice tests. The current study examined the classification performance of answer-copying indices measured by the area under the receiver operating characteristic (ROC) curve under different item response theory (IRT) models (one- [1PL], two- [2PL], three-parameter [3PL] models, nominal response model [NRM]) using both simulated and real response vectors. The results indicated that although there is a slight increase in the performance for low amount of copying conditions (20%), when nominal response outcomes were used, these indices performed in a similar manner for 40% and 60% copying conditions when dichotomous response outcomes were utilized. The results also indicated that the performance with simulated response vectors was almost identically reproducible with real response vectors.
Project description:Selection of an appropriate item response model is critical in the measurement of latent examinee ability. The one-, two-, and three-parameter logistic (1PL, 2PL, and 3PL) models are nested, and as such can be compared using likelihood ratio (LR) tests. The null hypothesis in the LR test for selection among the 2PL and 3PL models sets the guessing parameters to their lower bound of 0. This violates one of the assumptions of the LR test and renders the usual ?2 reference distribution inappropriate for the comparison. A review of the current literature revealed that this problem is not well understood in the educational measurement field. Ignoring this issue can lead to selection of an overly simplified model, with implications for the ability estimates. In this article, the use of the LR test for item response model selection is investigated, with the goal of providing practitioners with an appropriate method of selecting the most parsimonious model. The results of simulation studies indicate the nature of the problem, with inaccurate Type I error rates for cases where the inappropriate null distribution was used. An analysis of data from a statewide mathematics test showed differences pertinent to subsequent analyses.
Project description:The asymptotic power of the Mantel-Haenszel (MH) test for the differential item function (DIF) is derived. The formula describes the behavior of the power when the number of items is large, so that the measured latent trait can be considered as the matching variable in the MH test. As shown in the derived formula, the power is related to the sample size, effect size of DIF, the item response function (IRF), and the distribution of the latent trait in the reference and the focal groups. The formula provides an approximation of the power of the MH test in practice and thus provides a guideline for DIF detection in practice. It also suggests analytical explanations of the behavior of the MH test as observed in many previous simulation studies. Based on the formula, this study shows how to conduct the sample size calculation. The power of MH test under some practical models such as the two-parameter logistic (2PL) and three-parameter logistic (3PL) item response theory (IRT) models is discussed.
Project description:Due to the surge of interest in online retailing, the use of credit cards has been rapidly expanded in recent years. Stealing the card details to perform online transactions, which is called fraud, has also seen more frequently. Preventive solutions and instant fraud detection methods are widely studied due to critical financial losses in many industries. In this work, a Gradient Boosting Tree (GBT) model for the real-time detection of credit card frauds on the streaming Card-Not-Present (CNP) transactions is investigated with the use of different attributes of card transactions. Numerical, hand-crafted numerical, categorical and textual attributes are combined to form a feature vector to be used as a training instance. One of the contributions of this work is to employ transaction aggregation for the categorical values and inclusion of vectors from a character level word embedding model which is trained on the merchant names of the transactions. The other contribution is introducing a new strategy for training dataset generation employing the sliding window approach in a given time frame to adapt to the changes on the trends of fraudulent transactions. In the experiments, the feature engineering strategy and the automated training set generation methodology are evaluated on the real credit card transactions.
Project description:Unidimensional, item response theory (IRT) models assume a single homogeneous population. Mixture IRT (MixIRT) models can be useful when subpopulations are suspected. The usual MixIRT model is typically estimated assuming a normally distributed latent ability. Research on normal finite mixture models suggests that latent classes potentially can be extracted, even in the absence of population heterogeneity, if the distribution of the data is non-normal. In this study, the authors examined the sensitivity of MixIRT models to latent non-normality. Single-class IRT data sets were generated using different ability distributions and then analyzed with MixIRT models to determine the impact of these distributions on the extraction of latent classes. Results suggest that estimation of mixed Rasch models resulted in spurious latent class problems in the data when distributions were bimodal and uniform. Mixture two-parameter logistic (2PL) and mixture three-parameter logistic (3PL) IRT models were found to be more robust to latent non-normality.
Project description:The contrast of fraud in international trade is a crucial task of modern economic regulations. We develop statistical tools for the detection of frauds in customs declarations that rely on the Newcomb-Benford law for significant digits. Our first contribution is to show the features, in the context of a European Union market, of the traders for which the law should hold in the absence of fraudulent data manipulation. Our results shed light on a relevant and debated question, since no general known theory can exactly predict validity of the law for genuine empirical data. We also provide approximations to the distribution of test statistics when the Newcomb-Benford law does not hold. These approximations open the door to the development of modified goodness-of-fit procedures with wide applicability and good inferential properties.
Project description:Gold nanoparticles can be used as contrast agents for bio-imaging applications. Here we studied multi-photon luminescence (MPL) of gold nanorods (GNRs), under the excitation of femtosecond (fs) lasers. GNRs functionalized with polyethylene glycol (PEG) molecules have high chemical and optical stability, and can be used as multi-photon luminescent nanoprobes for deep in vivo imaging of live animals. We have found that the depth of in vivo imaging is dependent upon the transmission and focal capability of the excitation light interacting with the GNRs. Our study focused on the comparison of MPL from GNRs with two different aspect ratios, as well as their ex vivo and in vivo imaging effects under 760 nm and 1000 nm excitation, respectively. Both of these wavelengths were located at an optically transparent window of biological tissue (700-1000 nm). PEGylated GNRs, which were intravenously injected into mice via the tail vein and accumulated in major organs and tumor tissue, showed high image contrast due to distinct three-photon luminescence (3PL) signals upon irradiation of a 1000 nm fs laser. Concerning in vivo mouse brain imaging, the 3PL imaging depth of GNRs under 1000 nm fs excitation could reach 600 ?m, which was approximately 170 ?m deeper than the two-photon luminescence (2PL) imaging depth of GNRs with a fs excitation of 760 nm.
Project description:The consumption of raw fish has increased considerably in the West, since it is said to be potentially healthier than processed fish (for containing omega 3 and 6, essential amino acids and vitamins). However this potential benefit, as well as the taste, value and even the risk of extinction are not the same for all species of fish, constituting grounds for fraud. Using the principles of the DNA barcode we revealed mislabelling of fish in Japanese restaurants and fishmarkets in Florianópolis, a popular tourist capital in Brazil. We sequenced the COI gene of 65 samples from fisheries and 80 from restaurants and diagnosed 30% of mislabeled samples in fisheries and 26% in restaurants. We discussed that frauds may have occurred for different reasons: to circumvent surveillance on threatened species; to sell fish with sizes smaller than allowed or abundant species as being a much rarer species (law of supply); to induce product consumption using species with better taste. It should be noted that some substitutions are derived from incorrect identification and are not a fraud per se; they are due to confusion of popular names or misunderstanding by the sellers. Therefore, we suggest the implementation of a systematic regulatory program conducted by governmental agencies to reduce mislabelling in order to avoid further damage to the community (in health and financial issues) and fish stocks.
Project description:A three-parameter logistic (3PL) model variant, named the two-parameter logistic extension (2PLE) model, was developed. This new model employs a function that integrates item features according to an examinee's ability level instead of a fixed guessing parameter used in the 3PL model to quantify guessing behavior. Correct response probabilities from a solution behavior and guessing behavior increase as the level of ability increases. At extreme cases in which the level of ability is close to negative infinity, the 2PLE model degenerates into a 3PL model with a guessing probability at chance level (i.e., 1/m, where m is the number of options). The properties of the 2PLE model were described and compared with those of other guessing models. Then, a simulation study comparing the performance of the 2PLE model with that of the 3PL model under three scenarios was conducted. Results showed that the 2PLE model generally outperforms the 3PL model. Finally, the application of the new model in comparison with several existing models was demonstrated by using two real data sets.
Project description:BACKGROUND: Despite the importance of health care fraud and the political, legislative and administrative attentions paid to it, combating fraud remains a challenge to the health systems. We aimed to identify, categorize and assess the effectiveness of the interventions to combat health care fraud and abuse. METHODS: The interventions to combat health care fraud can be categorized as the interventions for 'prevention' and 'detection' of fraud, and 'response' to fraud. We conducted sensitive search strategies on Embase, CINAHL, and PsycINFO from 1975 to 2008, and Medline from 1975-2010, and on relevant professional and organizational websites. Articles assessing the effectiveness of any intervention to combat health care fraud were eligible for inclusion in our review. We considered including the interventional studies with or without a concurrent control group. Two authors assessed the studies for inclusion, and appraised the quality of the included studies. As a limited number of studies were found, we analyzed the data using narrative synthesis. FINDINGS: The searches retrieved 2229 titles, of which 221 full-text studies were assessed. We found no studies using an RCT design. Only four original articles (from the US and Taiwan) were included: two studies within the detection category, one in the response category, one under the detection and response categories, and no studies under the prevention category. The findings suggest that data-mining may improve fraud detection, and legal interventions as well as investment in anti-fraud activities may reduce fraud. DISCUSSION: Our analysis shows a lack of evidence of effect of the interventions to combat health care fraud. Further studies using robust research methodologies are required in all aspects of dealing with health care fraud and abuse, assessing the effectiveness and cost-effectiveness of methods to prevent, detect, and respond to fraud in health care.
Project description:Antisteatotic effects of omega-3 fatty acids (Omega-3) in obese rodents seem to vary depending on the lipid form of their administration. Whether these effects could reflect changes in intestinal metabolism is unknown. Here, we compare Omega-3-containing phospholipids (krill oil; ?3PL-H) and triacylglycerols (?3TG) in terms of their effects on morphology, gene expression and fatty acid (FA) oxidation in the small intestine. Male C57BL/6N mice were fed for 8 weeks with a high-fat diet (HFD) alone or supplemented with 30 mg/g diet of ?3TG or ?3PL-H. Omega-3 index, reflecting the bioavailability of Omega-3, reached 12.5% and 7.5% in the ?3PL-H and ?3TG groups, respectively. Compared to HFD mice, ?3PL-H but not ?3TG animals had lower body weight gain (-40%), mesenteric adipose tissue (-43%), and hepatic lipid content (-64%). The highest number and expression level of regulated intestinal genes was observed in ?3PL-H mice. The expression of FA ?-oxidation genes was enhanced in both Omega-3-supplemented groups, but gene expression within the FA ?-oxidation pathway and functional palmitate oxidation in the proximal ileum was significantly increased only in ?3PL-H mice. In conclusion, enhanced intestinal FA oxidation could contribute to the strong antisteatotic effects of Omega-3 when administered as phospholipids to dietary obese mice.