Characterize scales of measurement
(nominal, ordinal, continuous)
Interpret frequency distribution in
describing a set of biological
measurements
Define mean, median, mode,
percentile, variance, standard
deviation, range, and interquartile
range and describe how they are used
Distinguish between the standard
deviation and the standard error and
explain the appropriate use of each
Contrast the features of a normal
distribution with those of a skewed
distribution
Distinguish when to use
nonparametric vs. parametric
statistics
Explain what is meant by probability

Scales of measurement
o
Nominal, ordinal,
interval, ratio scales
o
Interval and ratio also
called numerical or
continuous
o
Possible properties of a
scale:
Identity:
each
number has an
inherent meaning
Magnitude:
inherent
order from small to
large
Equal intervals:
differences between
units on the scale is
the same, i.e. 1 to 2 is
same as 101 to 102
Absolute/true zero:
Zero represents
absence

Nominal Scales
o
A nominal scale is used
where arbitrary numbers
or designations can be
given to categories of the
data
o
Examples are gender, ethnic groups, disease state (i.e. hypertension, yes
or no)
o
No ordering is implied, i.e. 1=female, 2=male is same as 1=male,
2=female; male is not > female
o
No quantitative information, qualitative in nature
o
Identity property is only one that applies

Ordinal Scales
o
A scale that incorporates inherent order in the data
o
Properties include magnitude and identity, but NOT equal distances or
true zero
o
Examples include:
Cancer stages, i.e. lung cancer, stage I is confined to the lung,
stages II and III are confined to the chest, and IV is metastasis
Apgar scores, scale 010
Ranked preferences, i.e. soda or coffee
Levels of agreement, i.e. Likert scale

Interval Scale
o
A scale in which a given distance along the scale means the same
regardless of where on the scale
o
Properties include identity, magnitude, and equal distance, but NOT
true zero point
o
Differences make sense, but ratios do not
Examples include:
•
Temperature, Celsius or Farenheit
•
Dates
•
Test scores, i.e. SAT or GRE scores

Ratio Scale
o
Sa m e a s a n interval scale, but zero has m e a ning, n a m ely
the abs enc e of a characteristic
o
So properties include: identity, m a gnitude, equ al
o
It m akes s ens e to s ay 2 m is twice a s large as 1 m
o
Examples include:
Height, weight, length
Money
Biomarkers, i.e. creactive protein, cholesterol,
Sum m arizing Data

What is a frequ ency distribution?
o