Transcript Statistics: An Introduction - Blogs @ Suffolk University

Statistics: An Introduction
Alan Monroe: Chapter 6
Why Levels of Measurement are Important
Levels of measurement are important because they tell us (and
any computer program you might use) the specific values we
have assigned for certain data.
Most computers will interpret numerical values as interval, but
you might be defining or using it as nominal or ordinal data.
What is a Statistic? (88)
Numerical Measurement that Summarizes Data
It is a “numerical measurement that summarizes some
characteristic of a larger body of data. That is why statistics
are useful. They can reduce very large amounts of
information…to single numbers that convey information you
need.”
Examples of Statistics:
Totals: total population, total for dinner
Proportion: could be a fraction, a percentage
Rates: miles per gallon
Average (or mean)
Univariate Statistics (90)
Measures of Central Tendency: Averages
Mean:
The mean is computed by adding up all of the individual
values and dividing by the number of cases. Can only be
computed for Interval Data.
Median:
Median is the middle: “half cases have higher values and half
have lower values.” Often used to calculate income.
Mode: (Applies to Nominal, Ordinal, and Interval)
It refers to the “most frequently occurring value or category.
Univariate Statistics (91)
Measures of Dispersion:
It is a measure of “how closely or widely cases are separated on
a variable.”
Examples:
Range: the difference between the highest and the
lowest.
Standard Deviation: is based on the summation of the
difference of each case from the mean.
The Concept of Relationships (92)
Contingency Tables: (Cross-Tabulation)
This is a table showing the frequencies of each combination of
categories on the two variables. Can be used to present
nominal (party, gender, age) and ordinal data.
Scattergrams: (Scatterplots)
Relationships between two interval variables are shown in
scatterplots. Enables you to present interval data.
…
Multivariate Statistics (98)
Strength of a Relationship:
How good of a predictor is the IV of the DV. What is the
strength of the association? Is there a correlation between the
two variables?
Direction of a Relationship:
Is it a positive or negative relationship? When the IV increases,
does the DV increase (positive) or decrease (negative)?
Significance:
“probably caused by something other than mere chance
<statistically significant correlation between vitamin
deficiency and disease>.”
source: http://www.m-w.com/dictionary/significant