Descriptive Statisitcs Flashcards
What are the types of Frequency Distributions?
- Histogram
- Polygons
- Bar Graphs
Frequency distribution
Distribution showing a frequency for each value of a variable.
what is a frequency?
A count of how many times a value occurs
Nominal
NO: #s are not meaningful
NO: spaces between #s ≠
NO: |0|
Ex: gender, hair color
Ordinal
YES: #s are meaningful
NO: space between #s are ≠
NO: |0|
Organizing categories
Interval Data
YES: #s are meaningful
YES: spaces between #s are =
NO: |0|
Ratio
YES: #s are meaningful
YES: spaces between #s are =
YES: |0|
Zero point
Positively skewed data
high % of scores are at the low end
tail to the right
M is higher than the Mdn
Negatively Skewed Data
Large % of scores at the high end of distribution. Tail to the left.
M is lower than the Mdn
If Mean, Median, & Mode are equal
Symmetrical, no skew, both sides are equal, normal bell shaped curve.
Kurtosis
The degree to which scores cluster around the mean or are spread out over a wide range with many extreme scores.
It is a measure of the flatness or peakedness of a distribution.
Mesokurtic
Rolling hill.
Symmetrical, mesokurtic is a normal curve bell curve.
M, Mdn, Mode are equal
Platykurtic
Plato
Flat top, many scores at the extreme ends.
A lot of scores around the M but of equal purportionatley. a lot of scores at extreme ends
Leptokurtic
Peak or summit
Long tails, scores bunch up at the Mean.
M is Mode,
Measures of Central Tendency
Most typical or representative of the entire group.
Descriptive Statistics versus Inferential Statistics
Describes the variables
Inferential Infers about the relationship between variables
Mode
Most frequently occurring scores.
5, 10, 1, 1, 1
1 = Mode
Bimodal
Two scores occur at the highest frequencies.
Multi Model
All have to occur with equal frequency
Median
The score that divides the distribution directly in half where 50% of the scores are about it and 50% below it.
Odd # data middle most score.
Even # data at 2 middle scores and divide by 2
Mean
M = the sum of all individual scores
divided by n: total scores
EX = 40
EX/N = 40/4 = 10
N = all scores in a population
n = all scores of a different group
Mode:
Can be used with all scales of measurement.
ONLY one that can be used with NOMINAL DATA
Median
Can be used for Ordinal, Interval, and Ration Data
MOST appropriate for ORDINAL DATA
Special Considerations on the Mean
Measure of central tendency generally favored by psychologists
•Outliers: M is heavily influenced by extreme scores
Ex: how many sex partners would you like to have the next 30 years?
Measures of Variation
Range: subtract the highest score and lowest score
100-1 = 99
Variance, var, s2: # that represents the total amount of variation in the distribution, Larger the variance the more total spread of scores.
s2=Sum of all (take each score subtract from the mean, square it, then add it up.) divided by n-1.
not usually reported.
Standard Deviation:
how much on average scores deviate from the mean.
variance's square root. where: X= each score M = the mean or average n= the number of values E= means we sum across the values
What makes the SD big or small?
relative to what you are measuring. if the average age in class is 22 years old, sd is 0 means all ages are 22.
Mean
Can be used for Interval and Ratio Data but NOT for ORDINAL OR NOMINAL DATA
Ex: gold, silver, bronze cant have 3.5
Bar graphs
There are spaces between the bars indicate spaces between #s are ≠. Appropriate for Nominal or Ordinal data.
Appropriate distributions to be used with Interval and Ratio data?
Histogram
Polygon
