The thesis considers the development of teachers' assessment practice at key stage one from 1989 to 1995 as the National Curriculum assessment of the core subjects was piloted and implemented. It takes a longitudinal case study approach in two Hampshire Infant schools with additional evidence from other schools and LEA assessment advisers and places it within historical, research and personal contexts. It identifies how teachers' assessment and recording practice changed, how far these changes were temporary or sustained and what impact these changes had upon supporting children's learning. It proposes that in order for change to be implemented and sustained teachers' perceptions as well as their practice need to change. It compares teachers' perceptions of the relative importance of statutory tests/tasks with teacher assessment over the six years. The research shows that teachers of Y2 children were making increasing use of a range of methods for teacher assessment including observation and became increasingly confident in the dependability of their judgements through whole school planning and moderation activities. It found changes in science and design school planning and moderation activities. It found changes in science and design and technology assessment practice were not sustained when external requirements changed. It considers how schools assimilated and implemented changes in their practice and found that for change originating from one context (policy makers) and implemented in a separate context (teachers in schools) there was a time lag. National Curriculum assessment was subject to significant modifications over the six years. The thesis explores the nature of these changes, finding many were crisis led and often stimulated knee-jerk responses which had unplanned knock on effects. The research found that the absence of two way professional communication between originators and implementors limited the rate of change and the development of a common perception of teacher assessment and its significance compared to statutory tests and tasks. The evidence collected throughout the six years supports the importance of teachers' perception in the process of change and its effective implementation. It proposes that for effective implementation teachers should engage in a professional dialogue with policy makers to bring about evolutionary change in which consideration is given to the purpose and evaluation/research basis of the change required as well as the practice implications.

<div>In the U.S., unitary residential air conditioners are rated using standard AHRI 210/240 that is inadequate to credit equipment with advanced controls and variable-speed components since the ratings are based on results of steady-state laboratory tests. Contrarily, a load-based testing and rating approach is presented in this work that can capture equipment performance with its integrated controls and thermostat responses that is more representative of the field. In this approach, representative building sensible and latent loads are emulated in a psychrometric test facility at different indoor and outdoor test conditions utilizing a virtual building model. The indoor test room conditions are continuously adjusted to emulate the dynamic response of the virtual building to the test equipment sensible and latent cooling rates and the equipment dynamic response is measured. Meanwhile, the inlet temperatures to the test equipment thermostat are independently controlled to track the same virtual building response using a thermostat environment emulator that encloses the test thermostat, that provides typical flow conditions and of which the design and control are presented in this work. Climate-specific cooling seasonal performance ratings can be determined by propagating load-based test results through a temperature-bin method to estimate a seasonal coefficient of performance (SCOP). In addition, a next-generation rating approach is developed that extends load-based testing for performance mapping, such that the SCOP can be obtained using building simulations that incorporate specific building types, climates and an equipment-specific performance map. </div><div> </div><div>In this work, the proposed approaches were implemented to test and rate a variable-speed residential heat pump operating in cooling mode. Trained with results from only 12 load-based test intervals carried out using the test equipment, a quasi-steady-state mapping model was able to map the equipment performance across almost the entire operating envelope within $\pm10\%$ errors and the $R^2$ values were very close to 1. Using the identified performance map, the next-generation SCOP was obtained based on an annual simulation deployed in EnergyPlus, where the map was coupled to a typical single-family building in Albuquerque,NM. Compared to the temperature-bin-based rating, this simulation-based rating is able to comprehensively and appropriately reflect equipment annual field performance associated with a specific building type and climate, as the rating is extended from automated laboratory load-based testing and performance mapping.</div>