Brief Description
Pupils conduct a survey to investigate the ages of cars. They
are encouraged to discover why the sample may be biased and are
led to investigate official published figures to enable them to
answer other questions about cars.
Design time: 4-5 hours
Aims and Objectives
On completion of this unit pupils should be able to conduct a
simple counting survey and represent the results on a bar chart.
They will have practised working with published data, reading
tables, drawing histograms, calculating means and drawing simple
inferences from tabulated data.
The pupils should be more aware of possible bias in sampling;
how data are collected in different circumstances; some of the
problems that arise in connection with data collection, sources
of published data; and the difficulties that can arise in
interpretation and comparison of statistics.
Prerequisites
Pupils need to be able to (i) convert fractions to decimals, (ii)
plot points for a time series, (iii) use tally marks, (iv) draw a
bar chart and (v) work with percentages and change fractions to
decimals and percentages.
The optional section involves calculating a mean of a
frequency distribution. It will help if they have met this
technique before.
Equipment and Planning
Section A2 involves carrying out a survey of the last
letters of car number plates, and this will need planning. Page R1
is used to record the results. The unit starts with a discussion
of some of the questions that people may want to know about cars
and uses the results of the survey to estimate some answers.
Since the local survey is bound to be biased, as is shown by
comparing the estimates from the local survey with the national
figures available, time is taken to look at the effect of bias.
Section C uses national figures to look at the number of
new cars produced and to calculate the mean age of cars on the
road and the scrapping rate. A final optional section estimates
the age distribution of scrapped cars and calculates the mean
life of a car.
Section B4 is optional and gives practice in
estimation calculations first introduced in B1. C5
is a harder optional section for brighter pupils only. Individual
harder questions are marked with an asterisk.
Calculators would be useful for Sections B and C.
It is possible to interrupt the unit at the end of Section B
and return later to Section C.
Detailed Notes
Section A
A1
The questions are designed to initiate a discussion. To
find out how long cars last, the age at which the car is scrapped
is required. The growth rate of the total number of cars can be
determined from the total numbers licensed each year. A direct
method is to use published data (in the Monthly Digest of
Statistics). There is a time lag between the collection of
data and publication (perhaps as much as two years). Records are
now being kept centrally at the Driver and Vehicle Licensing
Centre, Swansea. Answers can also be estimated from surveys which
provide current data more quickly, but great care is needed, as
shown in this unit.
A2
The system of indicating the year by the end
registration letter should be discussed with the pupils. It was
changed in 1967 after pressure from car manufacturers to boost
mid-year sales. There should be fewer E registered than D
registered cars, but this may not show up, because of small
numbers. The letters I, and Q were omitted to
avoid confusion with the numbers 1 and 0. Q
is used for cars which have been registered but on which no tax
has yet been paid.
To carry out the survey, each group of three pupils will need
page R1 and an aid for counting up to 200 (eg. tallying or a list
of numbers from 1 to 200). Decide in advance where to take the
survey. There should be a reasonable traffic flow. It may be
possible to carry out the survey for homework. If there is little
traffic, results could be combined over several carefully chosen
time-periods (to avoid double-counting). It is important that
pupils carry out a survey to enable them to become aware of the
ensuing problems. The pupils should be reminded about road safety
prior to the survey. A car park could be used for further data.
Only private cars and vans are considered in the survey,
because other vehicles, such as lorries or buses, have a
different life-span. In 1975 the average annual distance
travelled by cars was 8600 miles; buses and coaches travelled
26800 miles, whilst goods vehicles travelled 13700 miles (Transport
Statistics 1965-75). Cars with foreign registration plates are
not included because their age is not obvious. Cars with 'personalized'
number plates are usually too few to have any significant effect,
except perhaps at some rally: one would get new cars showing old
registration plates.
A3
The result s given in Table 2 are to help pupils analyse
their own results. The survey was carried out at 9 a.m. on one of
the main roads leading into Rochdale. The distribution has few T
registered cars because the survey was conducted at the end of
August. The first class in the table 'F and earlier' may
include some personalized plates. The first class in the pupils'
table could be a later letter than F, depending on the
current letter.
A4
Care is needed when combining separate samples. Results
can only be combined if one can be fairly confident that a car
will not have been included more than once. Perhaps only some
results can be combined.
Section B
B1
The pupils calculate an estimate of the N
registered cars in Britain. Part e will not
apply if combined class results are used. Pupils should be made
aware that all estimates are only approximate. These particular
estimates will also be biased because of the method of sampling
used.
Some pupils might find it easier to work with fractions (such
as 17/200) and cancel rather than with
decimals.
B2
The results of the analysis clearly show the bias in the
Rochdale sample. Further calculations for other letter
registrations can be done.
B3
Pupils should write down their own answers to the
questions, but a final class discussion would set these in
perspective.
At 8 a.m. a higher proportion of newer company vehicles might
be noticed. On early closing day the bias would be against the
cars of local shoppers and/or the second car of the family (which
may be older). These groups might feature more strongly in a
sample on the town car park, which is less dependent on vehicle-traffic-hours.
There would be a different distribution in a strategic highly
priced park than a long-stay cheaper (or free) one, or the school
car park. A sample outside a secondhand car centre would be
biased towards older cars, while that in an area of expensive
houses might contain a higher proportion of newer cars. On an
industrial estate there are likely to be many business cars,
which will be of recent registration. These would not feature to
such an extent in a survey done on a Sunday. On motorways,
representative of longer, faster journeys, newer cars would tend
to predominate.
*B4
The survey was carried out on the Friday of a Bank
Holiday weekend and provides the data for extra practice on
previous work in the section. The number of touring caravans in
Britain is an estimated figure, based on the numbers produced,
their life span and use. The work assumes that touring caravans
are not left on any one site or that those left on sites over the
season will be towed by cars of the same age distribution. These
factors can be discussed as sources of bias. Estimates can be
obtained for other registration letters.
Section C
C1
Discussion on listing large numbers in terms of
thousands in tables may be beneficial. A common error in reading
such tables is to forget that the units are thousands. The early
questions are simple, to help pupils famaliarize themselves with
the table. They may require help in plotting the axis in e.
A more subtle point in the analysis of the time series is that
the proportion of new cars can appear to fall even when the
number of cars is still increasing, because it is expressed as a
fraction of a growing fleet. Economic factors may also be brought
into the discussion of trends. After a time there will probably
be an upturn, because cars have a limited lifespan.
C2
Parts a and b may be
omitted by the more able pupils. The column 'Average age' in
Table 5 may need explaining. The rationale for the last part of
the histogram may be expanded upon - it assumes an even
distribution of cars over the 4-year period. The data are plotted
(from 0 to 20 years) on the histogram in an opposite direction to
that of the bar chart in A3 (from 'F and
earlier' to current letter). If the sample were representative
and the present distribution similar to that in 1977, one would
get a mirror image.
If pupils have not used the mean of a distribution before,
part f may be omitted (see Cutting it Fine
or Seeing is Believing for work on the mean). The pupils
may need help in calculating the mean.
The formula may be
mentioned to brighter pupils who have already met the notation.
C3
The growth in the car population has been levelling off.
This could be because of saturation or a poor economy, or both.
Growth is calculated as a proportion of stock at the beginning of
the year (and similarly in the next section for cars scrapped).
Weaker pupils may prefer to omit part c.
C4
The scrapping rate has fluctuated over the years - it is
also affected by economic upswing or depression. Weaker pupils
may prefer to omit parts b and c.
*C5
This part may be found difficult by some pupils. All may
require further explanation of Table 6. There are a number of
hidden assumptions in the calculations, necessitated by lack of
data. The year of the first registration is given only for two-year
periods. The average ages taken in 1973 and 1975 are the mid-points
of the class intervals. The results are based on a sample first
taken in 1973, so it cannot include all the cars registered in
that year; hence these have been omitted. The average age at
scrapping is also the mid-point of an interval; a car registered
in December 1972, but scrapped in 1973 could be '0' years old,
while a car registered in January 1971, but scrapped in December
1974, would be 4 years old. 2 is taken as the mid-point.
The data refer only to private cars.
The histogram gives a visual impression of the lifetime of
cars. The mean age is more difficult to ascertain but provides a
definite figure.
Answers
A1 |
|
See detailed notes. |
|
|
|
A2 |
|
See detailed notes. |
|
|
|
A3 |
|
See detailed notes. |
|
|
|
B1 |
f |
See detailed notes. |
|
|
|
B2 |
c |
See detailed notes. |
|
|
|
B3 |
|
See detailed notes. |
|
|
|
B4 |
a |
51200 |
|
b |
12800 |
|
c |
112000 |
|
|
|
C1 |
c |
8247000, 11515000, 14047000 |
|
d |
1972, |
|
e |
1964, |
|
f |
1975 |
|
|
|
C2 |
a |
2006 thousand; 7, |
|
b |
1260 thousand, 11 |
|
c |
18 years; 137000 |
|
e |
See detailed notes. |
|
f |
(2125), (8289), 14395, 14042, (16533),
13860, 8606, 3510, (4932).
Total 86292 (thousand car-years)
Mean age = 86292/14040 = 6.14 (6) years |
|
|
|
C3 |
a |
655000, 780000, 142000, 108000 |
|
b |
1975 |
|
c |
0.0543, 5.4%, 0.0613, 6.1%, 0.0105, 1.1%,
0.0079, 0.8% |
|
|
|
C4 |
a |
1008000, 865000, 1092000, 1059000 |
|
b |
0.0836, 8.4%, 0.068, 6.8%, 0.0809, 8.1%,
0.0776, 7.8% |
|
c |
Smaller |
|
|
|
C5 |
a |
(11), 82, 172, (442), 651, 364, (512) |
|
b |
Mean age = 23988/2234 = 10.7 years |
References
Monthly Digest of Statistics (Central Statistical
Office)
Transport Statistics in Great Britain (Dept. of
Transport)
Vital Statistics about the Caravan Industry (National
Caravan Council)
Page R1
Registration letter |
Tally
marks |
Total |
No letter |
|
|
A |
|
|
B |
|
|
C |
|
|
D |
|
|
E |
|
|
F |
|
|
G |
|
|
H |
|
|
J |
|
|
K |
|
|
L |
|
|
M |
|
|
N |
|
|
P |
|
|
R |
|
|
S |
|
|
T |
|
|
V |
|
|
W |
|
|
X |
|
|
Y |
|
|
Z |
|
|
Total
|
|
Date:
Time:
Place:
Test Questions
- Give one reason for conducting a vehicle survey.
- Table T1 shows the results of a car survey. It was taken
on a Friday morning, 10.30 a.m. to 11.30 a.m., on a city
ring road.
End registration letter |
None |
A |
B |
C |
D |
E |
F |
G |
H |
J |
K |
L |
M |
N |
P |
R |
S |
Total |
Number of cars |
4 |
1 |
0 |
2 |
1 |
2 |
2 |
3 |
6 |
9 |
15 |
24 |
26 |
29 |
30 |
32 |
14 |
200 |
Table T1 - Car survey results, 20
January, 1978
- How many cars had no end letter?
- Which was the first letter to have more
registration than this?
- Copy and complete:
Half the cars were registered in the year of the
letter _____ or later.
-
- What is the advantage of adding results from
surveys done in different places?
- One group of pupils did a car survey at one end
of a road; another group did a survey at the
other end of the road. They combined their
results. Why might this be misleading?
-
- What is the fraction of H-registered
cars in Table T1?
- There were approximately 15 million cars licensed
in January 1978. Use your answer to 4a
to estimate the number of these that were H-registered.
- Notice when and where the survey in question 2
was carried out.
- Which of the following would you expect to
feature most in the survey?
Shopping traffic
Business cars
Holiday traffic
New deliveries to garages
How would you expect the results to differ:
- On a free town centre car park, again on Friday
morning?
- On a coast road during a Sunday afternoon in
summer?
- Table T2 gives the number of new buses compared to all
buses licensed that year.
Year |
Licences
current
(thousands) |
First
registered |
Number |
Percentage |
1965 |
81.7 |
5474 |
6.7 |
1966 |
78.5 |
5399 |
6.9 |
1967 |
78.8 |
5007 |
6.4 |
1968 |
79.7 |
5135 |
6.4 |
1969 |
79.1 |
5134 |
6.5 |
1970 |
77.8 |
5018 |
6.4 |
1971 |
18.1 |
6213 |
8.0 |
1972 |
76.7 |
6440 |
8.4 |
1973 |
78.8 |
7177 |
9.2 |
1974 |
78.6 |
5220 |
6.6 |
1975 |
79.6 |
5481 |
6.9 |
Table T2 - Buses licensed and first
registered 1965-75
(Source: Transport Statistics, Great Britain, 1965-75)
- In which years were new buses more than 7% of the
total licensed that year?
- In which year were new buses less than 6'5% of
the total licensed that year?
- How many buses were there in 1966; in 1967?
- How many more buses were there in 1967 than in
1966?
- Write the increase in buses in 1967 (from 1966)
as a fraction of the number of buses in 1966 (use
your answers to c and d)
- How many more buses were there in 1975 than in
1974?
- How many new buses were there in 1975?
- How many buses were scrapped from 1974 to 1975?
- What fraction of the 1974 buses were scrapped
between 1974 and 1975?
- A survey of shoppers is made between 10.30 and 11.30 on a
Tuesday morning. The sample contains 85% women. What
difference would you expect to find in a survey taken
during the same hours on a Saturday?
Answers
1 |
|
To determine the ages of vehicles and related ideas,
e.g. scrapping rate, growth rate, in the area where the
survey is carried out |
|
|
|
2 |
a |
4 |
|
b |
H |
|
c |
N |
|
|
|
3 |
a |
To increase the sample size |
|
b |
Most of the cars would have been included twice; the
surveys are ngt independent. |
|
|
|
4 |
a |
3/100 (6/200) |
|
b |
450000 |
|
|
|
5 |
a |
Business cars |
|
b |
More shoppers' cars, which would tend to be older
vehicles |
|
c |
Fewer business vehicles; relatively older cars would
be on the road |
|
|
|
6 |
a |
1971, 1972, 1973 |
|
b |
1967, 1968, 1970 |
|
c |
78500, 78800 |
|
d |
300 |
|
e |
3/785 |
|
f |
1000 |
|
g |
5481 |
|
h |
4481 |
|
i |
4481/78600 |
|
|
|
7 |
|
It would contain a smaller percentage of women (and a
greater percentage of men). |
Connections with Other Published Units from the Project
Other Units at the Same Level (Level 3)
Net Catch
Phoney Figures
Cutting it Fine
Pupil Poll
Multiplying People
Units at Other Levels In the Same or Allied Areas of the Currlculum
Level 1
If at first...
Tidy Tables
Leisure for Pleasure
Level 2
Opinion Matters
Level 4
Figuring the Future
Retail Price Index
Sampling the Census
Equal Pay
This unit is particularly relevant to: Social Sciences,
Mathematics.
Interconnections between Concepts and Techniques Used In these Units
These are detailed in the following table. The code number in
the left-hand column refers to the items spelled out in more
detail in Chapter 5 of Teaching Statistics 11-16.
An item mentioned under Statistical Prerequisites
needs to be covered before this unit is taught. Units which
introduce this idea or technique are listed alongside.
An item mentioned under Idea or Technique Used is not
specifically introduced or necessarily pointed out as such in the
unit. There may be one or more specific examples of a more
general concept. No previous experience is necessary with these
items before teaching the unit, but more practice can be obtained
before or afterwards by using the other units listed in the two
columns alongside.
An item mentioned under Idea or Technique Introduced
occurs specifically in the unit and, if a technique, there will
be specific detailed instruction for carrying it out. Further
practice and reinforcement can be carried out by using the other
units listed alongside.
Code No. |
Statistical
Prerequisites |
Introduced in |
2.1a |
Constructing single variable frequency
tables |
If at first...
Tidy Tables
Opinion Matters |
2.2a |
Bar Charts |
Leisure for Pleasure |
2.2j |
Plotting time series |
Cutting it Fine
Phoney Figures
Figuring the Future |
|
Idea or
Technique Used |
Introduced
in |
Also
Used in |
1.2a |
Using discrete data |
|
If at first...
Tidy Tables
Net Catch
Multiplying People
Figuring the Future
Retail Price Index
Leisure for Pleasure
Opinion Matters
Cutting it Fine
Phoney Figures
Sampling the Census
Equal Pay |
1.4b |
Using someone else's directly counted or
measured data |
Tidy Tables
Multiplying People
Sampling the Census |
Leisure for Pleasure
Figuring the Future
Equal Pay
Phoney Figures
Retail Price Index |
1.4e |
Finding appropriate data |
Tidy Tables |
Equal Pay |
3.1c |
Mean for small data set |
If at first...
Cutting it Fine
Sampling the Census |
Retail Price Index |
3.1f |
Mean for frequency distribution |
Cutting it Fine |
Equal Pay |
5x |
Comparing actual with expected values |
Figuring the Future |
If at first... |
Code No. |
Idea or
Technique Introduced |
Also Used in |
1.2c |
Problems of data classification |
Leisure for Pleasure
Phoney Figures
Equal Pay
Tidy Tables
Pupil Poll
Opinion Matters
Sampling the Census |
1.3b |
Sampling from a large population |
Net Catch
Pupil Poll
Retail Price Index |
1.3e |
Variability in samples |
If at first...
Pupil Poll
Net Catch
Cutting it Fine |
1.3h |
Biased samples |
Net Catch
Pupil Poll |
2.2g |
Histogram for grouped data |
|
3.1a |
Mode for discrete data |
Leisure for Pleasure
Equal Pay
Phoney Figures
Sampling the Census |
5a |
Reading tables |
If at first...
Opinion Matters
Phoney Figures
Equal Pay
Leisure for Pleasure
Net Catch
Figuring the Future
Tidy Tables
Multiplying People
Retail Price Index |
5b |
Reading bar charts histograms pie charts |
Leisure for Pleasure
Multiplying People
Tidy Tables
Phoney Figures
Cutting it Fine |
5c |
Reading time series |
Leisure for Pleasure
Phoney Figures
Cutting it Fine
Figuring the Future
Multiplying People |
5i |
Estimating population figures from
samples |
Net Catch
Pupil Poll
Retail Price Index |
5k |
Variability of estimates |
Seeing is Believing
Pupil Poll
Figuring the Future |
5u |
Inference from bar charts |
If at first...
Multiplying People
Phoney Figures |
5v |
inference from tables |
Leisure for Pleasure
Multiplying People
Sampling the Census
Tidy Tables
Phoney Figures
Retail Price Index
Net Catch
Figuring the Future
Equal Pay |
5z |
Detecting trends |
Cutting it Fine
Sampling the Census
Multiplying People
Equal Pay
Phoney Figures |
|