Understanding evaluation of learning support in mathematics and statistics

With rapid and continuing growth of learning support initiatives in mathematics and statistics found in many parts of the world, and with the likelihood that this trend will continue, there is a need to ensure that robust and coherent measures are in place to evaluate the effectiveness of these initiatives. The nature of learning support brings challenges for measurement and analysis of its effects. After briefly reviewing the purpose, rationale for, and extent of current provision, this article provides a framework for those working in learning support to think about how their efforts can be evaluated. It provides references and specific examples of how workers in this field are collecting, analysing and reporting their findings. The framework is used to structure evaluation in terms of usage of facilities, resources and services provided, and also in terms of improvements in performance of the students and staff who engage with them. Very recent developments have started to address the effects of learning support on the development of deeper approaches to learning, the affective domain and the development of communities of practice of both learners and teachers. This article intends to be a stimulus to those who work in mathematics and statistics support to gather even richer, more valuable, forms of data. It provides a 'toolkit' for those interested in evaluation of learning support and closes by referring to an on-line resource being developed to archive the growing body of evidence. © 2011 Taylor & Francis.

Ingroup and role identity influences on the initiation and maintenance of students' voluntary attendance at peer study sessions for statistics

Background: Although class attendance is linked to academic performance, questions remain about what determines students’ decisions to attend or miss class. Aims: In addition to the constructs of a common decision-making model, the theory of planned behaviour, the present study examined the influence of student role identity and university student (in-group) identification for predicting both the initiation and maintenance of students’ attendance at voluntary peer-assisted study sessions in a statistics subject. Sample: University students enrolled in a statistics subject were invited to complete a questionnaire at two time points across the academic semester. A total of 79 university students completed questionnaires at the first data collection point, with 46 students completing the questionnaire at the second data collection point. Method: Twice during the semester, students’ attitudes, subjective norm, perceived behavioural control, student role identity, in-group identification, and intention to attend study sessions were assessed via on-line questionnaires. Objective measures of class attendance records for each half-semester (or ‘term’) were obtained. Results: Across both terms, students’ attitudes predicted their attendance intentions, with intentions predicting class attendance. Earlier in the semester, in addition to perceived behavioural control, both student role identity and in-group identification predicted students’ attendance intentions, with only role identity influencing intentions later in the semester. Conclusions: These findings highlight the possible chronology that different identity influences have in determining students’ initial and maintained attendance at voluntary sessions designed to facilitate their learning.

Implementing an enhanced repository statistics system for QUT ePrints

Queensland University of Technology (QUT) was one of the first universities in Australia to establish an institutional repository. Launched in November 2003, the repository (QUT ePrints) uses the EPrints open source repository software (from Southampton) and has enjoyed the benefit of an institutional deposit mandate since January 2004. Currently (April 2012), the repository holds over 36,000 records, including 17,909 open access publications with another 2,434 publications embargoed but with mediated access enabled via the ‘Request a copy’ button which is a feature of the EPrints software. At QUT, the repository is managed by the library.QUT ePrints (http://eprints.qut.edu.au) The repository is embedded into a number of other systems at QUT including the staff profile system and the University’s research information system. It has also been integrated into a number of critical processes related to Government reporting and research assessment. Internally, senior research administrators often look to the repository for information to assist with decision-making and planning. While some statistics could be drawn from the advanced search feature and the existing download statistics feature, they were rarely at the level of granularity or aggregation required. Getting the information from the ‘back end’ of the repository was very time-consuming for the Library staff. In 2011, the Library funded a project to enhance the range of statistics which would be available from the public interface of QUT ePrints. The repository team conducted a series of focus groups and individual interviews to identify and prioritise functionality requirements for a new statistics ‘dashboard’. The participants included a mix research administrators, early career researchers and senior researchers. The repository team identified a number of business criteria (eg extensible, support available, skills required etc) and then gave each a weighting. After considering all the known options available, five software packages (IRStats, ePrintsStats, AWStats, BIRT and Google Urchin/Analytics) were thoroughly evaluated against a list of 69 criteria to determine which would be most suitable. The evaluation revealed that IRStats was the best fit for our requirements. It was deemed capable of meeting 21 out of the 31 high priority criteria. Consequently, IRStats was implemented as the basis for QUT ePrints’ new statistics dashboards which were launched in Open Access Week, October 2011. Statistics dashboards are now available at four levels; whole-of-repository level, organisational unit level, individual author level and individual item level. The data available includes, cumulative total deposits, time series deposits, deposits by item type, % fulltexts, % open access, cumulative downloads, time series downloads, downloads by item type, author ranking, paper ranking (by downloads), downloader geographic location, domains, internal v external downloads, citation data (from Scopus and Web of Science), most popular search terms, non-search referring websites. The data is displayed in charts, maps and table format. The new statistics dashboards are a great success. Feedback received from staff and students has been very positive. Individual researchers have said that they have found the information to be very useful when compiling a track record. It is now very easy for senior administrators (including the Deputy Vice Chancellor-Research) to compare the full-text deposit rates (i.e. mandate compliance rates) across organisational units. This has led to increased ‘encouragement’ from Heads of School and Deans in relation to the provision of full-text versions.

Fusing loop detector and probe vehicle data to estimate travel time statistics on signalized urban networks

This article presents a methodology that integrates cumulative plots with probe vehicle data for estimation of travel time statistics (average, quartile) on urban networks. The integration reduces relative deviation among the cumulative plots so that the classical analytical procedure of defining the area between the plots as the total travel time can be applied. For quartile estimation, a slicing technique is proposed. The methodology is validated with real data from Lucerne, Switzerland and it is concluded that the travel time estimates from the proposed methodology are statistically equivalent to the observed values.

Who claims gifts as a tax deduction? An examination of tax deductible gift statistics

Many donors, particularly those contemplating a substantial donation, consider whether their donation will be deductible from their taxable income. This motivation is not lost on fundraisers who conduct appeals before the end of the taxation year to capitalise on such desires. The motivation is also not lost on Treasury analysts who perceive the tax deduction as “lost” revenue and wonder if the loss is “efficient” in economic terms. Would it be more efficient for the government to give grants to deserving organisations, rather than permitting donor directed gifts? Better still, what about contracts that lock in the use of the money for a government priority? What place does tax deduction play in influencing a donor to give? Does the size of the gift bear any relationship to the size of the tax deduction? Could an increased level of donations take up an increasing shortfall in government welfare and community infrastructure spending? Despite these questions being asked regularly, little has been rigorously established about the effect of taxation deductions on a donor’s gifts.

Earned value management insights using inferential statistics

Purpose – The purpose of this paper is to summarise a successfully defended doctoral thesis. The main purpose of this paper is to provide a summary of the scope, and main issues raised in the thesis so that readers undertaking studies in the same or connected areas may be aware of current contributions to the topic. The secondary aims are to frame the completed thesis in the context of doctoral-level research in project management as well as offer ideas for further investigation which would serve to extend scientific knowledge on the topic. Design/methodology/approach – Research reported in this paper is based on a quantitative study using inferential statistics aimed at better understanding the actual and potential usage of earned value management (EVM) as applied to external projects under contract. Theories uncovered during the literature review were hypothesized and tested using experiential data collected from 145 EVM practitioners with direct experience on one or more external projects under contract that applied the methodology. Findings – The results of this research suggest that EVM is an effective project management methodology. The principles of EVM were shown to be significant positive predictors of project success on contracted efforts and to be a relatively greater positive predictor of project success when using fixed-price versus cost-plus (CP) type contracts. Moreover, EVM's work-breakdown structure (WBS) utility was shown to positively contribute to the formation of project contracts. The contribution was not significantly different between fixed-price and CP contracted projects, with exceptions in the areas of schedule planning and payment planning. EVM's “S” curve benefited the administration of project contracts. The contribution of the S-curve was not significantly different between fixed-price and CP contracted projects. Furthermore, EVM metrics were shown to also be important contributors to the administration of project contracts. The relative contribution of EVM metrics to projects under fixed-price versus CP contracts was not significantly different, with one exception in the area of evaluating and processing payment requests. Practical implications – These results have important implications for project practitioners, EVM advocates, as well as corporate and governmental policy makers. EVM should be considered for all projects – not only for its positive contribution to project contract development and administration, for its contribution to project success as well, regardless of contract type. Contract type should not be the sole determining factor in the decision whether or not to use EVM. More particularly, the more fixed the contracted project cost, the more the principles of EVM explain the success of the project. The use of EVM mechanics should also be used in all projects regardless of contract type. Payment planning using a WBS should be emphasized in fixed-price contracts using EVM in order to help mitigate performance risk. Schedule planning using a WBS should be emphasized in CP contracts using EVM in order to help mitigate financial risk. Similarly, EVM metrics should be emphasized in fixed-price contracts in evaluating and processing payment requests. Originality/value – This paper provides a summary of cutting-edge research work and a link to the published thesis that researchers can use to help them understand how the research methodology was applied as well as how it can be extended.

Regulating bereavement : inquests, family pressure and the gate keeping of suicide statistics

This study of English Coronial practice raises a number of questions, not only regarding state investigations of suicide, but also of the role of the Coroner itself. Following observations at over 20 inquests into possible suicides, and in-depth interviews with six Coroners, three main issue emerged: first, there exists considerable slippage between different Coroners over which deaths are likely to be classified as suicide; second, the high standard of proof required, and immense pressure faced by Coroners from family members at inquest to reach any verdict other than suicide, can significantly depress likely suicide rates; and finally, Coroners feel no professional obligation, either individually or collectively, to contribute to the production of consistent and useful social data regarding suicide—arguably rendering comparative suicide statistics relatively worthless. These issues lead, ultimately, to a more important question about the role we expect Coroners to play within social governance, and within an effective, contemporary democracy.

Nonparametric semantic segmentation for 3D street scenes

In this paper we propose a method to generate a large scale and accurate dense 3D semantic map of street scenes. A dense 3D semantic model of the environment can significantly improve a number of robotic applications such as autonomous driving, navigation or localisation. Instead of using offline trained classifiers for semantic segmentation, our approach employs a data-driven, nonparametric method to parse scenes which easily scale to a large environment and generalise to different scenes. We use stereo image pairs collected from cameras mounted on a moving car to produce dense depth maps which are combined into a global 3D reconstruction using camera poses from stereo visual odometry. Simultaneously, 2D automatic semantic segmentation using a nonparametric scene parsing method is fused into the 3D model. Furthermore, the resultant 3D semantic model is improved with the consideration of moving objects in the scene. We demonstrate our method on the publicly available KITTI dataset and evaluate the performance against manually generated ground truth.

Econometric Modelling with Time Series : Specification, Estimation and Testing

This book provides a general framework for specifying, estimating, and testing time series econometric models. Special emphasis is given to estimation by maximum likelihood, but other methods are also discussed, including quasi-maximum likelihood estimation, generalized method of moments estimation, nonparametric estimation, and estimation by simulation. An important advantage of adopting the principle of maximum likelihood as the unifying framework for the book is that many of the estimators and test statistics proposed in econometrics can be derived within a likelihood framework, thereby providing a coherent vehicle for understanding their properties and interrelationships. In contrast to many existing econometric textbooks, which deal mainly with the theoretical properties of estimators and test statistics through a theorem-proof presentation, this book squarely addresses implementation to provide direct conduits between the theory and applied work.

Familial typical migraine: Linkage to chromosome 19p13 and evidence for genetic heterogeneity

Migraine is a frequent familial disorder that, in common with most multifactorial disorders, has an unknown etiology. The authors identified several families with multiple individuals affected by typical migraine using a single set of diagnostic criteria and studied these families for cosegregation between the disorder and markers on chromosome 19, the location of a mutation that causes a rare form of familial hemiplegic migraine (FHM). One large tested family showed both cosegregation and significant allele sharing for markers situated within or adjacent to the FHM locus. Multipoint GENEHUNTER results indicated significant excess allele sharing across a 12.6- cM region containing the FHM Ca2+ channel gene, CACNL1A4 (maximum nonparametric linkage Z score = 6.64, p = 0.0026), with a maximum parametric lod score of 1.92 obtained for a (CAG)(n) triplet repeat polymorphism situated in exon 47 of this gene. The CAG expansion did not, however, appear to be the cause of migraine in this pedigree. Other tested families showed neither cosegregation nor excess allele sharing to chromosome 19 markers. HOMOG analysis indicated heterogeneity, generating a maximum HLOD score of 3.6. It was concluded that Chr19 mutations either in the CACNL1A4 gene or a closely linked gene are implicated in some pedigrees with familial typical migraine, and that the disorder is genetically heterogeneous.

Evaluating the use of online synchronous communication to enhance learning in statistics

According to social constructivists, learners are active participants in constructing new knowledge in a social process where they interact with others. In these social settings teachers or more knowledgeable peers provide support. This research study investigated the contribution that an online synchronous tutorial makes to support teaching and learning of undergraduate introductory statistics offered by an Australian regional university at a distance. The introductory statistics course which served as a research setting in this study was a requirement of a variety of programs at the University, including psychology, business and science. Often students in these programs perceive this course to be difficult and irrelevant to their programs of study. Negative attitudes and associated anxiety mean that students often struggle with the content. While asynchronous discussion forums have been shown to provide a level of interaction and support, it was anticipated that online synchronous tutorials would offer immediate feedback to move students forward through ―stuck places.‖ At the beginning of the semester the researcher offered distance students in this course the opportunity to participate in a weekly online synchronous tutorial which was an addition to the usual support offered by the teaching team. This tutorial was restricted to 12 volunteers to allow sufficient interaction to occur for each of the participants. The researcher, as participant-observer, conducted the weekly tutorials using the University's interactive online learning platform, Wimba Classroom, whereby participants interacted using audio, text chat and a virtual whiteboard. Prior to the start of semester, participants were surveyed about their previous mathematical experiences, their perceptions of the introductory statistics course and why they wanted to participate in the online tutorial. During the semester, they were regularly asked pertinent research questions related to their personal outcomes from the tutorial sessions. These sessions were recorded using screen capture software and the participants were interviewed about their experiences at the end of the semester. Analysis of these data indicated that the perceived value of online synchronous tutorial lies in the interaction with fellow students and a content expert and with the immediacy of feedback given. The collaborative learning environment offered the support required to maintain motivation, enhance confidence and develop problemsolving skills in these distance students of introductory statistics. Based on these findings a model of online synchronous learning is proposed.

Learning detectors quickly with stationary statistics

Computer vision is increasingly becoming interested in the rapid estimation of object detectors. The canonical strategy of using Hard Negative Mining to train a Support Vector Machine is slow, since the large negative set must be traversed at least once per detector. Recent work has demonstrated that, with an assumption of signal stationarity, Linear Discriminant Analysis is able to learn comparable detectors without ever revisiting the negative set. Even with this insight, the time to learn a detector can still be on the order of minutes. Correlation filters, on the other hand, can produce a detector in under a second. However, this involves the unnatural assumption that the statistics are periodic, and requires the negative set to be re-sampled per detector size. These two methods differ chie y in the structure which they impose on the co- variance matrix of all examples. This paper is a comparative study which develops techniques (i) to assume periodic statistics without needing to revisit the negative set and (ii) to accelerate the estimation of detectors with aperiodic statistics. It is experimentally verified that periodicity is detrimental.

Cell image classification using histograms, higher order statistics and adaboost

A cell classification algorithm that uses first, second and third order statistics of pixel intensity distributions over pre-defined regions is implemented and evaluated. A cell image is segmented into 6 regions extending from a boundary layer to an inner circle. First, second and third order statistical features are extracted from histograms of pixel intensities in these regions. Third order statistical features used are one-dimensional bispectral invariants. 108 features were considered as candidates for Adaboost based fusion. The best 10 stage fused classifier was selected for each class and a decision tree constructed for the 6-class problem. The classifier is robust, accurate and fast by design.

Discussion of : constructing summary statistics for approximate Bayesian computation: semi-automatic approximate Bayesian computation

This is a discussion of the journal article: "Construcing summary statistics for approximate Bayesian computation: semi-automatic approximate Bayesian computation". The article and discussion have appeared in the Journal of the Royal Statistical Society: Series B (Statistical Methodology).

Bluetooth vehicle trajectory by fusing Bluetooth and loops : motorway travel time statistics

Loop detectors are widely used on the motorway networks where they provide point speed and traffic volumes. Models have been proposed for temporal and spatial generalization of speed for average travel time estimation. Advancement in technology provides complementary data sources such as Bluetooth MAC Scanner (BMS), detecting the MAC ID of the Bluetooth devices transported by the traveller. Matching the data from two BMS stations provides individual vehicle travel time. Generally, on the motorways loops are closely spaced, whereas BMS are placed few kilometres apart. In this research, we fuse BMSs and loops data to define the trajectories of the Bluetooth vehicles. The trajectories are utilised to estimate the travel time statistics between any two points along the motorway. The proposed model is tested using simulation and validated with real data from Pacific motorway, Brisbane. Comparing the model with the linear interpolation based trajectory provides significant improvements.