MODULE 1
Statistic
can be divided into two parts:
Descriptive: |
Sorting
large amounts of data in ways, which enable its main features to be seen
immediately. |
Inferential: |
A
small amount of data which has been collected (call a sample) can be used to
infer general conclusions about the total amount of similar data that exist
uncollected in the world (called the population). |
A Priori ’ |
|
Relative Frequency |
Measurement
after a large number of trials. |
Subjective Approach |
(Bayesian)
Degree of belief expressed as a probability. |
Ø
Symmetrical,
Ø
Bell
Shaped, ‘
Ø
Unimodal.
The Hump is the average of the variable.
The standard deviation is the spread of the
variable.
1) Ambiguity of Definition |
ie does average reflect mean, median, or
mode? (esp average) |
2) Pictorial Representation Misleading |
Is graph misleading? |
3) Sample Bias |
Data
collection, phrasing of survey questions, interviewer bias? Is
like being compared with like? |
4) Omissions |
What
is missing Is
additional information missing that should have been included that could have
changed the conclusion |
5) Logical Error |
Do the numbers measure directly what is
being studied? Are there causal relationships? A
strong associative relationship may not be casual |
6) Technical Error |
Lack
of understanding of math details. Have
statistical definitions/techniques/methods been properly applied. |
Who is providing the evidence? |
|
Where did the data come from? |
Are
measurements equal, scales the same too many decimals |
Does it pass the common sense test? |
|
Has one of the 6 Common Errors been committed? |
|