Exam 3 Material Flashcards
Null hypothesis (Ho)
A research perspective in which there is no true difference between the comparison groups
Accepted - there is NO difference
Rejected - there IS a difference
Nominal data
Dichotomous/binary; non-ranked; named categories
No magnitude/order
No consistency of scale
No rational/absolute zero
Ex. Male vs. Female / colors / ethnicities
Ordinal data
Ordered, rank-able categories
HAS magnitude/order
No consistency of scale
No rational/absolute zero
Ex: satisfaction surveys / mood levels / pain scale (though some researches see it as interval data)
Interval data
HAS magnitude/order
HAS consistency of scale
No rational/absolute zero (can have negative values)
Ex: temperature / height / specific number on a known scale
What is the difference between interval and ratio data?
Interval data has an arbitrary 0 - can have (-) numbers
Ratio data has an absolute 0 - cannot have (-) numbers; “0” represents the absence of a measured value; typical of physiological measurements like BP, height, etc.
Discrete vs. Continuous data sets
Discrete = Nominal and Ordinal data
Continuous = Interval/ratio data
Descriptive statistics
Non-comparative, simple description of various elements of a study’s data
Mean vs. Median vs. Mode
MEAN = average
MEDIAN = the middle of a data set
MODE = most frequent number in a data set
Ex. 1, 1, 1, 2, 4, 5, 5, 8, 8
Mean = (1+1+1+2+4+5+5+8+8)/9 = 3.88
Median = 4
Mode = 1
What do outliers affect?
The mean/average of a data set
Interquartile range
IQR = Q3 - Q1
The middle 50% of a range of data
Variance
The average of the squared differences between each individual measurement value and the groups’ mean
Standard deviation
Square root of the variance
Normal distribution
When the mean of a data set is equal (or nearly equal) to the median, such that there is equal dispersion of a curve’s tails on either side of the mean
Skewness = 0 Kurtosis = 0
What tests are useful for normally distributed data?
Parametric tests
Positive skew
When the mean > median for a data set
Results in a curve shifted to the LEFT (tail point to the right)