Download Descriptive Statistics

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts
no text concepts found
Transcript
Descriptive Statistics
I. Frequency Distributions
II. Measures of Central Tendency
III. Measures of Dispersion
I. Frequency Distributions
Frequency
– the number of observed cases for each value of the variable under observation
– EX:
N= 1,500
FEMALE:
MALE:
900
600
60%
40%
II. Measures of Central Tendency
A. MODE
– the value of a variable that is most frequently observed
B. MEDIAN
– the value of a variable that has 50% of the cases above it & 50% below it
C. MEAN
– the sum of the values for all cases divided by the number of cases --> “the average”
II. An Example
30
20
10
22
7
12
1
18
2
6
3
2
0
0
4
5
1
2
3
4
5
6
7
6
7
III. Measures of Dispersion
A. RANGE
– the difference between the highest observed value of a variable and the lowest
– in previous example
HIGH VALUE = 6
LOW VALUE = 1
RANGE = 5
– NOTE: Range doesn’t capture clusters of cases in middle or at extremes...
B. VARIATION
– sum of the squared deviations from the mean
C. VARIANCE
– variation divided by the number of cases
**for a sample population the denominator is N -1
D. STANDARD DEVIATION
– square root of the variance
III. (cont.) -> an example
x
x – x (x - x)2
4
-4
16
6
-2
4
8
0
0
9
1
1
10
2
4
11
3
9
TOTAL: 48
0
34
E. THE NORMAL DISTRIBUTION
– often called the “bell curve”
– the mean, median, & mode are identical
– a perfectly symmetrical distribution with most cases clustered around the mean
– 68% of the cases are within 1 standard deviation of the mean
– 95% are within 2 std. deviations
– 99.7% are within 3 std. deviations
A Normal Distribution
III. (cont.)
F. Z-SCORES
– a “raw score” expressed in standard deviations
– score’s deviation from the mean divided by the standard deviation
– useful for comparing scores (or measures) to rest of population of a nearly normal
distribution
– EX: Test Mean = 79.25; s = 7.19
SCORE = 95 --> 15.75 from mean
Z-SCORE = 2.19
Related documents