Statistics: Descriptive Stats Basics Review Flashcards
Continuous vs. Discrete Variables
Continuous = infinite number of values (e.g. time, age)
Discrete = limited set of values
Dichotomous = Discrete variable with only 2 values
(e.g. HS graduate or not)
Nominal scale
Categorical
Unordered
e.g. religion, political affiliation
Can ONLY measure frequency, nothing else
Ordinal Scale
Categorical
Ordered (ranked)
E.g. ranks and ratings on a Likert scale
Limitation: no way to determine amount of difference between scores – it is simply more than or less than
Interval Scale
Ordered
Equal intervals
No absolute zero
Limitation: Can only add and subtract without a zero
Ratio Scale
Ordered
Equal intervals
Absolute zero point
Allows for all mathematical operations
Descriptive statistics: overview
Describe or summarize the distribution (set) of data
Frequency Polygons
Data sets that represent ordinal, interval, or ratio scales can be organized in a frequency polygon
X axis = scores
Y axis = frequencies
Normal Distribution (Normal Curve)
Shape of frequency polygon shape when a sufficiently large number of observations are made
Symmetrical
Bell-shaped
Defined by specific mathematical formula
Kurtosis
Shape of the curve in terms of the relative peakedness of distribution
*normal curve = mesokurtic
Platykurtic Distribution
Distribution is flatter than normal curve
“Plat kurt, Flat Curve”
Leptokurtic Distribution
Distribution is more peaked than the normal curve
Measure of Central Tendencies for different Measurement Scales
Nominal: Mode
Ordinal: Mode or Median
Interval: Mode, Median, or Mean
Ratio: Mode, Median, or Mean
Adding a constant to every score in a distribution of scores will…
Increase the mean
But NOT effect of the SD
Distribution Percentile Rank compared to SD’s
M= 50%
1SD = 84%
2SD= 97.7%
3SD = 99.9%