Measures of central tendency Flashcards
1
Q
what are measures of central tendency?
A
- single values describing a set of data by identifying central position within that set of data
- mode
- median
- mean
2
Q
What is the mode?
A
- the value that occurs most frequently
- most useful when working with nominal variables (only useful measure of central tendency for nominal variables)
3
Q
What is the median?
A
- the middlemost value in a distribution
- always the exact centre of a distribution
- half of the scores in a distribution are higher than the median and half are lower than median
4
Q
How do you calculate the median when there’s an odd number of cases?
A
- all scores must be arranged in order from low to high
- when there are an odd number of cases the median is the exact middle case
eg. 10 15 25 40 60 > median is 25
5
Q
How do you calculate the median when there’s an even number of cases?
A
- all scores must be arranged in order from low to high
- Find case number by taking the number of scores, adding 1, then divide by 2 > (N+1) ÷2
- median is the value between the 2 middlemost cases
- Find the two middle cases, add them and divide by 2
eg. 10 15 25 40 60 70 > (25+40) ÷ 2 = 32.5
6
Q
What are limitations of the median?
A
- cannot be calculated for nominal variables
- can be found in ordinal or interval-ration variables
- generally preferred measure of central tendency for ordinal variables because mean can be misleading
7
Q
How is the mean calculated?
A
X̄= ( Σ xi ) / N
where:
X̄ is the sample mean
Σ xi (sigma) is the summation of all the scores
N is the total number of cases
8
Q
What is the sum of each observation’s deviation from the mean?
A
Å (Xi - X̄)
9
Q
Why is the median a better measure of central distribution than the mean?
A
- Mean can be misleading if the distribution is skewed
- because it uses all the scores in its calculation, it can be affected by unusually high/low scores
- mean is pulled in the direction of these extreme values: when few unusually high values, the mean is pulled higher; when few unusually high values, the mean is pulled lower
10
Q
What is skew?
A
- skew refers to when an otherwise normal distribution has a few extremely high or a few extremely low scores
- positively skewed: few high scores (pulled up to the left)
- negatively skewed: few low scores (pulled up to the right)