3. Descriptive Statistics:
Functions of Descriptive Statistics:
a) Data reduction:
allow a few numbers to summarize many
b) Measures of association
quantify strength and direction of a relationship.
Types of Descriptive Statistics:
Univariate Descriptive Statistics:
Describes the distribution of a single variable
Bivariate Descriptive Statistics
Summarize relationship between two variables
Multivariate Descriptive Statistics
Summarize relationship between 3 or more
variables
4. Inferential Statistics:
Functions of Inferential Statistics:
a) Aids researchers in drawing inferences from
samples to populations
Population
all cases in which the researcher is interested.
Samples
carefully chosen subsets of the population.
b) Used to test significance of hypotheses
5. Types Of Variables
Variables may be:
Independent or dependent
Discrete or continuous
Nominal, ordinal, or interval-ratio
6. Types Of Variables:
Independent or Dependent
In causal relationships:
CAUSE
EFFECT
independent variable dependent variable
7. Types Of Variables:
Discrete or Continuous
Discrete variables are measured in units that cannot
be subdivided.
Anything regarding people
Wrong when you hear the average number of children per
household being 2 ½…
Continuous variables are measured in a unit that
can be subdivided infinitely.
8. Types of Variables:
Level Of Measurement
Levels of Measurement represent different
levels of numerical information contained
within the variable:
1. Nominal
2. Ordinal
3. Interval-ratio
9. Nominal Level Variables
Used to classify or categorize
Simplest (or crudest) level of measurement
Numbers are for classification purposes only
Examples:
Religion: 1 = Protestant, 2 = Catholic, 3=Jew,
4=None, 5=Other
Gender (dichotomy): 1 = Female, 0 = Male
10. Ordinal Level Variables
Used to rank or order in a logical way
Scores can be ranked from high to low or from more
to less
Offers the property of “more than” or “less than” to
classifications
11. Example of an Ordinal Level Variable:
“Do you agree or disagree that University Health
Services should offer free contraceptives?”
5=strongly agree
4=agree
3=neutral
2=disagree
1=strongly disagree
Because you can distinguish between the
scores of the variable using terms such as
“more, less, higher, or lower” the variable is
ordinal.
12. Interval-ratio Variables
Scores are “actual” numbers
Ratio variables meet the criteria for interval but also
have meaningful and true zero points
Have equal intervals between scores indicating exact
distance, thus:
Can indicate “how much more” (or less)
Permits the use of mathematical operations
Examples:
Age (in years)
Income (in dollars)
Number of children
A true zero point (0 = no children)
Equal intervals: each child adds one unit
13. Different statistics require different
mathematical operations (ranking, addition,
square root, etc.)
The level of measurement of a variable tells us
which statistics are permissible and
appropriate.
15. Classify the level of measurement of the
following variables and tell level of
measurement:
The numbers on an athlete’s jersey
The dorm you live in
Number of children in a family
Fear of crime ( a lot, some, none)
Number of hours per day respondent watches tv
Tuition in dollars