The general aims of the project are to provide:

• a demonstrably improved quality of service to dissertation students, leading to similar improvement in dissertation content;
  
• a student reference tool for data 'mining' and subsequent qualitative and quantitative analyses;
  
• a benchmark standard against which to 'measure' student grades; this will be used to monitor student performance and thereafter provide 'auditable' information that will feed into quality control loops (and further refinement, where needed, to the internet tool).

Other incidental outcomes are also possible. Because statistical knowledge is a precursor for higher degrees (MPhil's and PhD's), undergraduate students who demonstrate an aptitude for exploiting data could be identified as potential higher degree candidates. In turn, supervisors of higher degree students could use the benchmark standard developed to assess potential candidates' academic performance.

It is recognised that the web page will not provide the definitive answer to all statistical principles, rather, an insight into statistics that will enable students to gain invaluable knowledge and experience which will enhance their dissertation submissions and prepare them better for industry.

Resources:

The project was funded (£3000) as a result of a competitive bid to the University's Learning and Teaching Strategy Project 2001/2

Staff time is required to maintain and enhance the web page on the basis of user feedback; subsequent funding is being sought for this.

Users must have web access. The Web page is deliberately not password protected or restricted to Wolverhampton university students' access. This is because the tool is intended to be freely and easily accessible by all. In turn, the potential for feedback and enhancement of the tool will (hopefully) be increased.

Details:

The synopsis outlines the difficulties that students of the Built Environment experience in coping with statistical methods particularly in interpreting data gathered in their projects. These problems are exacerbated by the fact that over fifty per cent of these students attend University on a part-time basis which limits their ability to access the library and search for appropriate text books.

Statistical analysis is not a requisite taught undergraduate subject in any course offered by the Built Environment division, however statistics performs an integral and quintessential role in decision-making and forecasting situations. Inherently, construction projects require a considerable amount of applied 'computational' mathematical and statistical knowledge which might be considered a precursor to student success within industry.

The challenge, therefore, was to provide students with an easily accessible (internet) statistical reference tool in the form of pragmatic 'construction biased' worked examples and additional supplementary information methods such as conversion of qualitative into quantitative data. These processes would help them learn fundamental software applications (for example SPSS, Minitab or Excel), how to interpret data generated thereafter and more importantly how such data relates to industry.

The stages in the development of the tool were as follows:

1. Determine which statistical methods were to be included in the Web page (and those to be excluded) and construct a data transformation web page. The process of selecting which statistical procedures were to be included in the Web page was based on a broad literature review of introductory statistical texts. From these a number of key statistical procedures were identified.
2. Based on the findings in 1, it was decided that the web page would include discursive statistical measurement; measurements of dispersion; probability; normal distribution; correlation; t-testing ; chi-squared testing and regression.
3. Test prototype web page using sample student group
4. Make revisions as necessary
5. Advertise web page to prospective students
6. Monitor and evaluate Web page use through user feedback and update as necessary.

At the time of writing the project has reached stage 4, where an initial evaluation (see below) has taken place leading to some reprogramming. Further, after testing the alternatives it was decided to use only Excel as the applications software package. This is because Excel is far more widely available than other packages and has a powerful statistical analysis component which is more than adequate for the needs of the project.

Evaluation:

Developmental Evaluation
To ensure that the web page contained sufficient detail for students to understand its content, a random sample of students was asked to comment on it. (random in this sense is defined as students from different Built Environment courses and at different stages of progression). Student feedback was, and is, an important element in the development of the web page as failure to communicate effectively with students will lead to it being rejected by those it is intended to help.

Initially, a paper survey was undertaken by forty students with, however, some reluctance to fill in the questionnaire based on several grounds. Firstly, students felt that this formal approach could identify an individual's potential mathematical weakness (and thus impact upon grades awarded); secondly, coursework and exam commitments intervened (which they perceived, quite rightly, as being of the utmost importance). To resolve this problem, face-to-face informal interviews were held with students, each being given strict assurances of confidentiality. Initial comments identified that the 'conceptual' idea of an interactive web page was of interest to students; though the actual page failed to spell out in sufficiently explicit detail, step-by-step instructions regarding data manipulation a priori to analysis.

Because of this critical observation, the project programme was adjusted so that greater detail (in the form of prescriptive step-by-step guidance augmented with screen dumps) can be included in the Web page to ensure that the students' interest is retained. This proved to be a challenging task and subsequently took considerably more time than was allowed for in the initial programme. Some reprogramming of the project has subsequently been carried out and the Web page has been made formally accessible to final year undergraduate students since Easter 2002. Whilst this additional work and time was unexpected at the outset, it has contributed to the production of this tool through a 'robust' iterative process of development and testing which should ensure that an optimum product is delivered to students.

Final Evaluation
Final evaluation of the tool will be made on four levels.

• initially, the quality of students' dissertations will be monitored to observe changes in data analysis methods used;
  
• secondly, use of the Web page will be monitored by having 'visitor counters' and 'cookies' inserted. Such information should provide a valuable insight into the frequency with which the pages are accessed, by whom and from which physical location;
  
• third, taking a long-term view, undergraduate candidates who have exploited any data they have gathered and have understood the statistical concepts in the web page, could be targeted as possible candidates for higher degrees;
  
• finally, a sample of students will be surveyed to identify what benefits they have gained from being able to refer to an easily accessible tool to help them with their studies. Potential areas this survey could concentrate upon are: the students' perception of how the tool has assisted their data analysis methods; how learning about statistical concepts has assisted them at job interviews and how the tool has developed student capabilities in a ways that were not anticipated.

Benefits:

As the web page has only been illustrated to a sample of students it is not yet possible to confirm long-term benefits. User feedback during the development stage however indicates that students are extremely positive about the statistical Web page, not least because statistics are applied in a contextual 'construction' background as against theoretical concepts that are often presented in various (so-called introductory) textbooks. Students believed the Web page would:

• improve their confidence in understanding and applying statistical concepts and data;
• enhance the quality of their dissertations by improved analysis of quantitative data;
• improve their IT skills;
• give user-friendly, easy access to a valuable resource while at university and during employment

Issues and Lessons Learned:

As explained earlier students struggle with understanding mathematical concepts, hence they are likely to shy away from using statistical analysis procedures. Generally many students also have poor time keeping skills. These issues suggest the need for learning tools to be quick and easy to use. Hence, the web page development which aims to 'get to the point' so that students can analyse their data easily and at the earliest opportunity.

It must be remembered, however, that the web page is intended as a 'taster' to illustrate to students that statistics are a useful tool. Its downside is that it comprises a 'bare bones' statistical text (as demanded by the students) in which there is great potential for over-simplifying a complex subject.

Once the web page is launched, it is envisaged that it will undergo continual revision based on feedback received so that potential criticism is minimised.

Ongoing Developments:

This type of resource is constantly being developed as new and better information becomes available. The author continuously reviews the teaching methods used for delivering this subject and attempts to make access increasingly more user-friendly.

References, Further Reading and Sources of Further Related Information:

For further information see:

1. Statistical Resources: a web based student edition of Minitab available at: www.belmont.edu/Humanities/Behavioral.Sci/PSY/PSY260/Stat.htm
   
2. Kent, R. (2001) Data construction and data analysis for survey research: Palgrave: Basingstoke.