Psychological Statistics Flashcards
Nominal Scale of Measurement
Name different categories
No quantitative properties
Numbers are just used as names
Ordinal Scale of Measurement
Numbers differentiate categories or instances
Numbers convey an ordered series of relationship
Can’t assume equal increments (1 to 2)
Example? Ranks
Interval Scale of Measurement
Has all properties of nominal and ordinal
A numerical difference means the same thing across the scale
Zero does not mean absence of thing measured
Example? degrees F, C, or
Ratio Scale of Measurement
Has all properties of nominal, ordinal, and interval
Zero point is absolute
Example? time, degrees K
discontinuous (discrete) variable
Variable for which only a finite number of values can occur between any two points
E.g. 1 person or 3 persons not 1.5 person or 2.7 person
Continuous variables
Theoretically can assume an infinite number of values between any two points
E.g. height, age, reaction times
independent variable
the variable manipulated by the researcher
dependent variable
the variable being measured by the researcher
Real (true) limits
applied to a continuous variable and it is (+/-) one half of the unit measured (decimal place). E.g.
47 + .5= 47.5
47 - .5= 46.5
or 47.5 + .05= 47.55
47.5 - .05= 47.45
Rounding Conventions
Round to two decimal places
Greater than 5 round up
Less than 5 leave unchanged
Exactly 5 round up to nearest even digit if not already there
6.575 –> 6.58
6.525 –> 6.52
Frequency Histogram
frequency polygon
Line Plot
Mesokurtosis
Platykurtosis
LeptoKurtosis
stem and leaf plot
Frequency Distributions for Quantitative Variables: Ungrouped Scores
scores are arranged from highest to lowest (x)
second column indicates the frequency (f)
detailed
Frequency Distributions for Quantitative Variables: Grouped Frequency Distribution
arranged from highest to lowest
specify score INTERVALS and count how many scores fall in each interval
less detailed
mode
most frequent score in a data set
data set can have 2 modes making it bimodal
median
middle score
if N is even (N+1)/2
take the average of adjacent scores
mean
X bar for sample
mu for population
positively skewed
mode - median - mean
negatively skewed
mean - median - mode
range
X highest - X lowest
Sensitive to outliers
interquartile range
difference between the 25th percentile and the 75th percentile
isn’t affected by outliers so much but does omit data
standard deviation
variance
