|
Crossing the Road - More by Judgement than Luck
Bob Neale
(Meldreth, England) (From The Best of Teaching Statistics)
This has been a three year project and my last three classes of year six children (aged 10- 1 1) have taken an active part in the testing, recording, data processing and analysis of the information we have collected. It all started during our school safety week. My class were doing their cycling proficiency training and road safety was the theme for the week for the whole school. I read my class an article, from their local newspaper, headlined "County's bleak child death toll" claiming that Cambridgeshire had one of the bleakest records, for child pedestrians killed on its roads, in Europe (Figure 1).
| We then discussed why this should be. As part of the discussion I told them that I live in a purpose built, new village where pedestrians and traffic are kept as separate as possible, and that several people have suggested that children from the village would find crossing roads, in busier places, difficult as they lack experience of traffic. We then discussed this and concluded that children from busy traffic environments would be able to judge the speed of oncoming traffic, when crossing the road, better than children from quieter traffic environments (Hypothesis 1). We decided that we should test this hypothesis and, at the same time, test the hypothesis that the Green Cross Code needs expanding to include judgement training (Hypothesis 2). I told the children to think about how we could go about testing our hypotheses. They quickly realised that we would need to devise a safe, fair, roadside test, that would not require children to actually cross the road. They organised themselves into groups and each group devised a test. We ended up with seven possibilities and by discussion and practical trialling finally decided upon a suitable but simple test that we could use anywhere. We carried out our testing in eight schools, two in busy city environments, two in busy rural environments, two in quiet rural environments, one in a 'medium' environment (nowhere near as busy as 'busy' but not as quiet as 'quiet') and one in a very quiet rural environment. Once the testing was complete we had a vast amount of data that we stored in the Grass Database program in our Acorn A3000 2Mb computer. We then analysed the data using the Grass program itself, the Data Sweet data handling package and, of course, pencil and paper methods. | ![]() |
| County's bleak child death toll
CAMBRIDGESHIRE has one of the bleakest records for the number of child pedestrians killed , it has been claimed. The county is amongst the worst in Europe even though the UK has the lowest road death rate in Europe, county road safety officers have revealed. Their special report shows pedestrian casualties in Cambridgeshire have increased from 349 in 1987, to 386 in 1988. And they have welcomed new Government proposals aimed at cutting the death toll across the country. County road safety officer Bob Pearson said: 'The increases are mainly in Cambridgeshire and in the small towns and villages around. "All accidents are tragic, but those happening to very young children trying to cross busy roads on their own are particularly so.' |
Gathering the statistics, and the mathematics involved in the initial
analysis, were straightforward enough but we had to make sense of this mass
of statistics and decide whether our hypotheses had been proved or not.
This was much harder. The children had made some very reasonable deductions from the raw statistics, as we collected them, but when we then percentaged them this made a nonsense of their original deductions. To begin with they found this hard to accept but quickly realised the need to use percentages and continued the analysis with renewed vigour. Gathering the statistics, and the mathematics involved in the initial analysis, were straightforward enough but we had to make sense of this mass of statistics and decide whether our hypotheses had been proved or not. This was much harder. The children had made some very reasonable deductions from the raw statistics, as we collected them, but when we then percentaged them this made a nonsense of their original deductions. To begin with they found this hard to accept but quickly realised the need to use percentages and continued the analysis with renewed vigour. |
Figure 1. Article from Royston Crow, 5 May 1989
They finally came to the conclusion that our first hypothesis was wrong as the children from the busy city environments, although very confident, were not as good at crossing the road safely as the children from the busy rural environments who were less confident but more careful and very accurate with their judgements. Table 1 and Figure 2 supply the evidence. We devoted a great deal of thought and discussion to our second hypothesis. It occurred to us that if Green Cross Code is sufficient training then the results more or less the same - but they weren't. They were in fact very different. So we considered that our second hypothesis was proved to be correct. Table 1 supplies the evidence. Finally we concluded that children from some traffic environments are better at judging speed, distance and time of oncoming traffic than children from other traffic environments, or to quote one of my children:- "I really thought that all children had the same ability to cross the road, but they haven't, have they?" Children need to be taught more than just the existing Green Cross Code as it should also include judgement training. Children also need extra training designed specifically for their traffic environment and the problems it poses them. My pupils learnt a considerable amount from this project. They learnt how to design a research project, the importance of a fair test and the importance of checking and rechecking the project design at each stage. They learnt how to collect, record, represent and analyse statistical data, and use a computer database. They also developed their abilities to use statistics in decision making and learnt the importance of accepting that the statistics can prove their hypotheses wrong! They also learnt the importance of taking care when crossing roads. "We went to Cambridge on Saturday and I told my parents off because they made us cross the road in a dangerous way." The project also made them think very carefully about their own safety when crossing the road or riding their bicycles, especially when not in their own familiar village environment, as one of the children said,"This has made me realise that 1 must be much from each of the traffic environments would have been more careful in Cambridge and Royston." Table 1
| Busy | City | Busy | Rural | Medium Rural | Quiet | Rural | Very Quiet Rural | |
| Item description | A | B | C | D | E | F | G | H |
| 1 Yes (would cross the road) (%) | 72 | 48 | 46 | 4 | 82 | 28 | 70 | 16 |
| 2 No (would not cross the road) (%) | 28 | 52 | 54 | 96 | 18 | 72 | 30 | 84 |
| 3 Decision in less than 2 seconds (%) | 72 | 55 | 65 | 77 | 77 | 65 | 95 | 96 |
| 4 Decision in more than 2 seconds (%) | 28 | 45 | 35 | 23 | 23 | 35 | 5 | 4 |
| 5 Run Over (if they had crossed) (%) | 19 | 14 | 0 | 4 | 0 | 0 | 0 | 8 |
| 6 Too cautious, could have crossed with 3 or more seconds to spare - but didn't (%) | 0 | 0 | 0 | 0 | 41 | 20 | 15 | 4 |
| 7 Made right decision (%) | 81 | 86 | 100 | 96 | 59 | 80 | 85 | 88 |
| 8 Would have been run over if said 'Yes': Percentage of 'No's (%) | 78 | 93 | 21 | 32 | 17 | 21 | 0 | 71 |
| 9 Near miss if said 'Yes'; Percentage of 'No's (%) | 11 | 0 | 21 | 24 | 6 | 0 | 0 | 0 |
| 10 Average Car Time (secs) | 8.16 | 8.03 | 6.61 | 6.03 | 8.5 | 7.88 | 7.51 | 6.97 |
Please email: alison.davies2@ntu.ac.uk with any comments or corrections.
©
The Teaching Statistics Trust 2006. The Teaching Statistics Trust is a registered
charity.
ISSN 0141-982X (Print) ISSN 1467-9639 (Online)