The computation of equating errors in international surveys in education

Since the IEA's Third International Mathematics and Science Study, one of the major objectives of international surveys in education has been to report trends in achievement. The names of the two current IEA surveys reflect this growing interest: Trends in International Mathematics and Science Study (TIMSS) and Progress in International Reading Literacy Study (PIRLS). Similarly a central concern of the OECD's PISA is with trends in outcomes over time. To facilitate trend analyses these studies link their tests using common item equating in conjunction with item response modelling methods. IEA and PISA policies differ in terms of reporting the error associated with trends. In IEA surveys, the standard errors of the trend estimates do not include the uncertainty associated with the linking step while PISA does include a linking error component in the standard errors of trend estimates. In other words, PISA implicitly acknowledges that trend estimates partly depend on the selected common items, while the IEA's surveys do not recognise this source of error. Failing to recognise the linking error leads to an underestimation of the standard errors and thus increases the Type I error rate, thereby resulting in reporting of significant changes in achievement when in fact these are not significant. The growing interest of policy makers in trend indicators and the impact of the evaluation of educational reforms appear to be incompatible with such underestimation. However, the procedure implemented by PISA raises a few issues about the underlying assumptions for the computation of the equating error. After a brief introduction, this paper will describe the procedure PISA implemented to compute the linking error. The underlying assumptions of this procedure will then be discussed. Finally an alternative method based on replication techniques will be presented, based on a simulation study and then applied to the PISA 2000 data.     

The influence of journal publisher characteristics on open access policy trends

Examines SHERPA/RoMEO publisher open access (OA) policy information for 100 publishers over a 13 year period (2004–2016) to consider whether their size, type or country (UK or US) affected the development of their OA policy over time. A publisher’s RoMEO colour code, whether they offered a Gold OA option, and the mean number of restrictions as to when, how and where papers may be self-archived, were all mapped. Kruskal–Wallis tests were run to assess whether the differences between their 2004 and 2016 positions were statistically significant. Finds that the growth of Green and Gold OA policy approaches has not been evenly distributed amongst publishers with some significant differences amongst publishers of different size, types and country (UK and US). Large commercial publishers are more likely to be allocated a RoMEO colour code, but at the same time place a high volume of restrictions as to where and how authors might self-archive. Small publishers are less likely to have a RoMEO green colour code, but the volume of restrictions they place on self-archiving are minimal. University presses appear not to be engaging with either OA agenda to any considerable degree. UK and US publishers’ OA policies appear to be influenced by the national OA policy environment which, considering the global nature of the scholarly journals market, was more pronounced than might have been anticipated.     

Test equating in the presence of DIF items

This paper proposes a multilevel measurement model that controls for DIF effects in test equating. The accuracy and stability of item and ability parameter estimates under the proposed multilevel measurement model were examined using randomly simulated data. Estimates from the proposed model were compared with those resulting from two multiple-group concurrent equating designs, including 1) a design that replaced DIF-items with items with no DIF; and 2) a design that retained DIF items with no attempt to control for DIF. In most of the investigated conditions, the results indicated that the proposed multilevel measurement model performed better than the two comparison models.     

The influence of safety belt laws on self- reported safety belt use in the United States

We assessed rates and trends in safety belt use by presence and type of safety belt law using data from states participating in the 1984–1989 Behavioral Risk Factor Surveillance System. State(s) with a safety belt law allowing law enforcement officers to stop vehicles for occupants' failure to use safety belts (primary enforcement law) had greater and more rapid increases in safety belt use rates than did states with laws requiring that vehicles must first be stopped for some other violation before a citation or fine for occupants' failure to use safety belts could be imposed (secondary enforcement law). Larger and sustained increases in safety belt use occurred when safety belt laws became effective or when fines were imposed for violations than when laws were first enacted. These data suggest that primary enforcement laws result in greater and more rapid increases in safety belt use than do secondary enforcement laws, and that initial increases in safety belt use following implementation of laws are sustained.     

离散密集频谱细化分析与校正方法研究进展

对近年来频谱校正领域中离散密集频谱细化分析与校正方法的理论研究和发展现状进行了回顾。根据信号所包含的频率分量的数目,将密集频谱细化分析与校正方法分为两类：一类是包含两个密集频率成分信号的频谱细化方法,另一类是包含三个及以上密集频率成分信号的频谱细化方法。综合阐明了各种离散密集频谱细化分析与校正方法的基本思想、算法原理、特点及其在工程中的应用。分析了现有密集频谱细化分析与校正方法的优缺点,并对离散密集频谱细化分析与校正领域的发展前景进行了展望。     

Evaluation of the 0.3 logits screening criterion in common item equating

A number of state assessment programs that employ Rasch-based common item equating procedures estimate the equating constant with only those common items for which the two tests' Rasch item difficulty parameter estimates differ by less than 0.3 logits. The results of this study presents evidence that this practice results in an inflated probability of incorrectly dropping an item from the common item set if the number of examinees is small (e.g., 500 or less) and the reverse if the number of examinees is large (e.g., 5000 or more). An asymptotic experiment-wise error rate criterion was algebraically derived. This same criterion can also be applied to the Mantel-Haenszel statistic. Bonferroni test statistics were found to provide excellent approximations to the (asymptotically) exact test statistics.     

An investigation of factors affecting test equating in latent trait theory

The study investigated five factors which can affect the equating of scores from two tests onto a common score scale. The five factors were: (a) item distribution type (i.e., normal versus uniform; (b) standard deviation of item difficulty (i.e.,.68,.95,.99); (c) number of items or test length (i.e., 50, 100, 200); (d) number of common items (i.e., 10, 20, 30); and (e) sample size (i.e., 100, 300, 500). SIMTEST and BIGSTEPS programs were used for the simulation and equating of 4,860 item data sets, respectively. Results from the five-way fixed effects factorial analysis of variance indicated three statistically significant two-way interaction effects. Simple effects for the interaction between common item length and test length only were interpreted given Type I error rate considerations. The eta-squared values for number of common items and test length were small indicating the effects had little practical importance. The Rasch approach to equating is robust with as few as 10 common items and a test length of 100 items.     

Comparing screening approaches to investigate stability of common items in Rasch equating

This paper used real and simulated data sets to compare three screening approaches often used in state-wide equating programs utilizing the Rasch model: Wright and Stone's t-statistic, robust z-statistic, and displace. Analyses of real data sets supported the superiority of robust z-statistic and displace measure relative to Wright and Stone's t-statistic. The simulation component did not support the contention that indiscriminate use of the +/-0.3 logits criterion inflates rates of Type I error for robust z-statistic and displace measure, although this contention was supported for the Wright and Stone's t-statistic. However, Type II error rates were largest for displace measure, followed by the robust z-statistic, then the t-statistic. The paper discusses the importance of a priori selection of a criterion for screening linking items and its effects on stability and accuracy of Rasch equating constant.     

A multidimensional Rasch analysis of gender differences in PISA mathematics

Since the 1970s, much attention has been devoted to the male advantage in standardized mathematics tests in the United States. Although girls are found to perform equally well as boys in math classes, they are consistently outperformed on standardized math tests. This study compared the males and females in the United States, all 15-year-olds, by their performance on the PISA 2003 mathematics assessment. A multidimensional Rasch model was used for item calibration and ability estimation on the basis of four math domains: Space and Shape, Change and Relationships, Quantity, and Uncertainty. Results showed that the effect sizes of performance differences are small, all below .20, but consistent, in favor of boys. Space and Shape displayed the largest gender gap, which supports the findings from many previous studies. Quantity showed the least amount of gender difference, which may be explained by the hypothesis that girls perform better on tasks that they are familiar with through classroom practice.     

How reinforcement sensitivity and perceived risk influence young drivers’ reported engagement in risky driving behaviors

Gray's reinforcement sensitivity theory (RST), implementing Carver and White's behavior inhibition system (BIS) and behavior approach system (BAS) scales, was used to predict reported engagement in 10 risky driving behaviors: speeding (2 levels), driving under the influence of alcohol, racing other vehicles, cell phone use (hand-held and hands free), tailgating, unsafe overtaking, driving while fatigued, and not wearing a seatbelt. Participants were 165 young male and female (n = 101) drivers aged 17–25 years who held a valid Australian driver's license. Effects of the explanatory variables and specific risk perceptions upon engagement in the reported risky driving behaviors were examined using SEM analyses. Also of interest was whether perceived risk mediated the relationship between the personality variables and reported engagement in risky driving behaviors. RST variables, negative reactivity, reward responsiveness and fun seeking, accounted for unique variance in young drivers’ perceived risk. Reward responsiveness and perceived risk accounted for unique variance in young drivers’ reported engagement in risky driving behaviors. Negative reactivity was completely mediated by perceived risk in its negative relationship with reported engagement. To better understand driving related risk decision making, future research could usefully incorporate drivers’ motivation systems. This has the potential to lead to more tailored approaches to identifying risk-prone drivers and provide information for the development and implementation of media campaigns and educational programs.     

The role of alcohol use on recent trends in distracted driving

Distracted driving is now an increasingly deadly threat to road safety. We provide evidence that intoxicated driving is increasingly responsible for recent increases in fatalities from distracted driving crashes. This study describes trends in deaths on U.S. public roads caused by alcohol-involved and distracted drivers using the Fatality Analysis Reporting System (FARS)—a census of fatal crashes on U.S. public roads. Fatality rates per vehicle-miles traveled are calculated using data from the Federal Highway Administration. Alcohol-involved drivers who are simultaneously distracted were responsible for 1750 deaths in 2009, an increase of more than 63% from 2005 when there were 1072 deaths. Alcohol use while driving is increasingly responsible for a growing number of fatalities from distracted driving, accounting for 32% of deaths from distracted driving in 2009 versus 24% in 2005. The fatality rate from these crashes increased from 35.9 to 59.2 deaths per 100 billion vehicle-miles traveled after 2005. Alcohol use is quickly increasing as an important factor behind distracted driving fatalities. This has implications for policies combating distracted driving that do not address the role of alcohol use in distracted driving.     

Review and analysis of publications on scientific mobility: assessment of influence,motivation, and trends

Scientometrics - The phenomenon of scientific mobility, actively developing in recent decades, attracts increasing attention of researchers in view of its importance for the development of science,...     

The influence of sputtering on FeSi

The effect of different (0.5, 2 and 4 keV) Ar⁺ energy ions on (a) the composition of an FeSi surface, (b) the oxidation of the FeSi surface after bombardment, and (c) the segregation of silicon after bombardment, has been monitored by Auger electron spectroscopy. Silicon was found to be preferentially sputtered by the Ar⁺ ions at all the different energies during bombardment. This effect was more pronounced at the 0.5 keV ion energy bombardment. There was a slight increase in the oxidation rate from the higher to the lower Ar⁺ ion energy at which the sample was sputtered before oxidation. The rate of silicon diffusion to the surface at 593 K after the sample had been sputtered, was lower when the sample had been sputtered by 0.5 keV ions than by 2 keV ions.     

The influence of wear on stresses in PCD inserts

The paper addresses the behavior of maximum stresses on the tool face and flank and the tool failure probability with increasing wear. The maximum tool life is calculated from the wear rate diagram. The authors provide some research findings on the influence of a Ni-resist cast iron embedded in the piston on the wear resistance of a PCD insert.     

溶液的杂质含量对单泡声致发光的影响

1引言: 在适当除气的液体中,单一气泡可以悬浮在超声驻波场的波腹处,并随声场的频率作周期性的膨胀和收缩,在收缩时将声能量高度集中,会产生发光的现象,称为单泡声致发光(Single Bubble Sonoluminecence)[1].这个现象的发现距今才十几年,还有许多地方需要探索和改进.     

The impact of publications from mainland China on the trends in alphabetical authorship

This paper investigates the impact of burgeoning Chinese publication on academic alphabetical authorship in the 25 subject categories that have the highest percentage of intentionally alphabetical publications. The use of alphabetical authorship is common in the social sciences and humanities, mathematics, and in some physical disciplines. Chinese academic publication has increased rapidly in recent decades (Hong Kong and Macau were excluded from the study because Hong Kong and Macau are much more internationalized than mainland China). However, authors from mainland China do not prefer alphabetical authorship. The increase in publications from mainland China lowers the probability of intentional alphabetical authorship in the natural science and technology subject categories that we examined. In some natural science and technology categories, the influence is strong. But for the social sciences and humanities, the influence is weak, due to the lower share of world publications from mainland China. Yet, in some social science and humanities subject categories such as ‘Economics’, the relative share of publications from mainland China is increasing rapidly, and the results on alphabetical authorship trends will be felt in the near future.     

The influence of scattering on magnetic breakdown

A study of magnetic breakdown between the third zone needles and second zone monster surfaces in the three zinc alloy systemsZnMn,ZnCr, andZnFe shows that the tunneling probability is strongly reduced by electron scattering and that the breakdown fieldH ₀ increases linearly with the Dingle temperatureX with a scaling factor of 0.36 kG/K for all alloys. This phenomenon may be understood if we allow the electrons to be scattered during their tunneling through the forbidden region ofk space. A simple semiclassical calculation of this effect yields an expression for the breakdown probability corrected from the conventional expression toP=exp[–(1/H)(H₀+X)], where is a constant dependent only on the band structure of the solvent.