John R. Davy

Student debt and its relation to student mental health

The present paper provides an analysis of the relationship between attitudes toward debt and mental health among university undergraduates. Data were collected from the same cohort of students across their three years of university, with responses from 2146, 1360 and 1391 first, second and third year students, respectively. Mental health was measured using the General Population version of the Clinical Outcomes in Routine Evaluation (GP-CORE). Attitudes toward debt were measured using items that tapped current financial concerns and worry about debt on leaving university. Results showed that students become more concerned about their finances as they progress through university, that there was no relationship between anticipated debt and mental health and that attitudes toward debt were related to mental health levels. Students who were identified as having high financial concerns possessed significantly worse CORE-GP scores than students with low financial concern in all three years of university. In all three years students with high financial concerns felt more ‘tense, anxious or nervous’, more ‘criticised by other people’ and found it more ‘difficult getting to sleep or staying asleep’ than students with low financial concerns. There was also evidence that students with high worry about their debt anticipated leaving university with higher amounts of debt than low debt worry students. These findings are discussed in relation to the pattern of increased student debt in UK higher education.

Diversity and Motivation in Introductory Programming

Abstract The number of students entering Higher Education in the UK continues to increase, and there is no sign of an end to this trend. This expansion brings with it a more diverse student body — students are diverse in terms of their prior experiences, their pre-existing skills, their expectations and their motivations. This highly varied student body often encounters a teaching regime that was designed for a smaller, and much more homogeneous, group of students. This is a significant problem in Computing courses, and especially in introductory programming. Students will approach learning to program from a wide variety of backgrounds, yet they will usually be taught and assessed in the same way. This paper considers the diversity of the introductory programming class, and describes some attempts to handle this diversity in the teaching programmes at the School of Computing at the University of Leeds.

Research-led innovation in teaching and learning programming

We describe an attempt to bridge the gap between educational research and practical innovation by making a package of changes to an introductory programming module based on the insights of existing theoretical work. Theoretical principles are described, used to evaluate previous practices and then employed to guide systematic changes. Preliminary evaluation indicates substantial improvements in student performance and enjoyment while indicating areas in need of further work.

Student well-being in a computing department

We describe a project exploring the relationships between factors in the learning environment, student well-being and learning outcomes, in the context of a Computing department. A range of established psychometric tests identified areas of unhelpful stress in the working environment and measures were implemented to rectify these. A significant improvement in measured student well-being followed.

Implementation Issues Relating to the WPRAM Model for Scalable Computing

Modern parallel processing machines are becoming more scalable through advances in network technology. It is now important to have a scalable computational model to support the design and analysis of algorithms. This paper describes a practical implementation of the WPRAM model, which has been used at Leeds for a number of years. The distinctive features of the WPRAM are the use of a weakly coherent shared address space, and the support of fine-grain and highly irregular forms of parallelism. The implementation strategy concentrates on the issues of scalability and good practical performance, with particular attention given to the support of the shared address space.

A Parallelisation Approach for Supporting Scalable and Portable Computing

This paper describes a strategy for the structuring and analysis of parallel code, using shared abstract data types (SADTs). SADTs are used to provide scalability, and support for modular and portable code development. An example of their usage is presented for a dynamic load balancing method. A framework for performance analysis is described, using an extension of the bulk synchronous parallelism (BSP) approach, and performance results are presented for the Cray T3D.

A polymorphic library for constructive solid geometry

Solid modelling using constructive solid geometry (CSG) includes many examples of stylised divide-and-conquer algorithms. We identify the sources of these recurrent patterns and describe a Geometric Evaluation Library (GEL) which captures them as higher-order functions. This library then becomes the basis of developing CSG applications quickly and concisely. GEL is currently implemented as a set of separately compiled modules in the pure functional language Hope+. We evaluate our work in terms of performance and general applicability. We also assess the benefits of the functional paradigm in this domain and the merits of programming with a set of higher-order functions.

The effect of varying computational granularity and tree imbalance on the performance of parallel tree-based application

In this paper we discuss the effect of two tree application parameters, namely computational granularity and tree imbalance, on the overall performance of parallel tree structured application. We apply a technique called phased-based adaptive dynamic load balancing, a version of adaptive DLB in which a parallel computation moves through different load balancing phases identified on the basis of run-time workloads. A simulator was used to evaluate the benefit of this approach. Two DLB algorithms were used as test cases; the Generalized Dimension Exchange Method and Load server algorithms. The results show that applications with large grain computation would benefit most from our approach while tree imbalance does not show consistent advantage.

Generating Parallel Applications of Spatial Interaction Models

We describe a tool enabling portable parallel applications of spatial interaction modelling to be produced automatically from high-level specifications of their parameters. The application generator can define a whole family of models, and produces programs which execute only slightly more slowly than corresponding hand-coded versions.

Capturing Branch-and-Bound using Shared Abstract Data-types

To support the routine construction of large-scale parallel applications requires an effective mechanism of abstracting from the underlying machine. In this paper, abstraction using Shared Abstract Data-types is illustrated through a case-study of an irregular problem; the Travelling Salesman Problem. This design of a Branch and Bound algorithm to solve this problem is investigated, demonstrating the separation of algorithmic and implementation issues that the SADT approach offers. Issues in the composition of SADTs, and methods of exploiting the shared data weakness/performance tradeoff are discussed.