3. FREQUENCY DISTRIBUTIONS Flashcards
1
Q
- What do we look at when describing the distribution of a numeric value?
A
- we look for the overall pattern
- we look for striking deviations from the pattern
2
Q
- What 3 characteristics do we describe a histogram by?
A
- its shape
- its centre
- its spread
3
Q
- What kind of Distribution is this?
A
- it is symmetric
- it is normal (Gaussian)
- it is unimodal
4
Q
- What kind of Distribution is this?
A
- it is skewed
- it is unimodal
- it is not symmetric
5
Q
- What kind of Distribution is this?
A
- this is a Bimodal Distribution
6
Q
- What does Unimodal mean?
A
- this means one peak
- this means that there is a single most common value
7
Q
- What does Bimodal mean?
A
- this means two peaks
8
Q
- What does Multimodal mean?
A
- this means multiple peaks
9
Q
- When is a distribution said to be Symmetrical?
A
- this is also known as a Normal Distribution
- this happens if the right and left sides of the Histogram
are approximately mirrored images of each other
10
Q
- What two other names can we call a Symmetrical distribution?
A
- a Bell Curve
- a Gaussian Distribution
11
Q
- What is a Positively Skewed Distribution?
A
- this is when a distribution is skewed to the right
- the right side of the Histogram extends much further
out than the left side - this means that the distribution is said to have a right
tail
12
Q
- What is a Negatively Skewed Distribution?
A
- this is when a distribution is skewed to the left
- the left side of the histogram extends much further out
than the right side - this means that the distribution is said to have a left
tail
13
Q
- What can be said about the Mode across these three different kinds of distributions?
A
- the mode is unchanged
- the graph shape has no effect on it
14
Q
- What can be said about the Median and the Mean across these three different kinds of distributions?
A
- the Median and the Mean are pulled towards the
direction of the skew - the results of the Median would be more reliable than
the results of the mean for this situation
15
Q
- What are the Mean and the Median for this Normal Distribution?
A
16
Q
- What are the Mean and the Median for this Positively Skewed Distribution?
A
- the Mean is pulled towards the skew more severely
than the Median - the Median reflected the data better
17
Q
- What parameters do you use when the data is skewed?
A
- you use the Median
- you use the IQR
- DO NOT USE the standard deviation
- DO NOT USE the mean
18
Q
- What is an Outlier?
A
- Outliers are observations that lie outside the overall
pattern of a distribution - make sure to always look for outliers and interpret
them
19
Q
- What happens to the original mean when we have an outlier?
A
- the mean is changed dramatically
- it is pulled towards the direction of the skew
20
Q
- What happens to the original median when we have an outlier?
A
- the median remains largely unchanged
21
Q
- Provide descriptions for Labels 1-4.
A
- Skewed Right
- Skewed Left
- Normally Distributed Variable
- Skewed Variable
22
Q
- Which methods should be use to check the distribution of numeric variables?
A
- a histogram
- a box-plot
23
Q
- For which kind of distributions should we only use Box-plots and histograms?
A
- for normal distributions
- this is because the mean is affected by skewness and
outliers - it should only be used for variables that do not have
outliers
24
Q
- Do outliers always affect the mean?
A
- NO
- a few outliers will not affect the mean of a very large
sample
25
Q
- Does skewness always effect the mean?
A
- YES
- no matter how small the sample is
26
Q
- What affects the Standard Deviation?
A
- skewness
- outliers
27
Q
- In which situations should we make use of the Standard Deviation?
A
- it should only be used when the variable is normally
distributed - it should only be used when there are no outliers
present
28
Q
- What is the only parameter that is infrequently used in Scientific Research?
A
- mode
29
Q
- What percentage of values falls into a sample contained in a range that is one standard deviation from the mean?
A
- 68%
NB:
- this applies explicitly for distributions that are perfectly
normal
30
Q
- What percentage of values falls into a sample contained in a range that is two standard deviations from the mean?
A
- 95%
NB:
- this applies explicitly for distributions that are perfectly
normal