Sie sind auf Seite 1von 15

Variation

Variation is a measure of the


differences between members of
the same species.
Where does variation come from?
• Genetic origin: human blood group, eye
colour, Cepea shell colour
• Environmental factors e.g. spoken
language, ability to play a musical
instrument
• Combination of both of these
• e.g. height, weight, intelligence,
Variation can be continuous or
discontinuous
• Continuous: no clear demarcation
between groups, such as height, hair
colour, weight. Any groupings are
calculated artificially.
• Discontinuous: each group is clearly
different, with no link between them, such
as blood group, gender, tongue-rolling
Normal Distribution
Many forms of variation follow a pattern
called the normal distribution.
A mean value is calculated from the data.
The frequency of each data value is
plotted on a graph.
This pattern is symmetrical about the
mean, and the data values nearest to the
mean have the highest frequencies.
So the graph has this shape:
Investigating Variation
Hypothesis: Americans are fatter than
Europeans.
There are two possible strategies to test
this hypothesis:
1. Weigh every American and every
European and compare the means of the
data.
2. Weigh some from each group (samples)
and test whether the samples are
significantly different.
Strategy 1 would take too much time and
be very difficult to complete, so Strategy 2
is the only reasonable option.

Samples must be taken in such as way as


to ensure randomness (avoiding bias)
and reliability (to reduce the effect of
anomalies).
If the Americans really are fatter than
Europeans, we would say that they come
from a different population, one which
has a mean weight that is significantly
higher than the mean weight of the
European population.
That conclusion would be fine if you really
had measured every American and every
European, but if you have taken a sample
there will be anomalies and extreme
results.
You need to be able to decide whether the
anomalies are
• part of the normal spread of the data,
• errors that have occurred as a result of faulty
technique or equipment
• or whether the difference is more significant
(and the original hypothesis may be wrong).
Standard Deviation
Standard deviation is a mathematically
calculated measure of the spread of the
data. Each data point is compared to the
mean value and the differences used to
calculate a single figure.
The greater the value of the standard
deviation, the greater the spread of the
data about the mean.

Das könnte Ihnen auch gefallen