Professional Documents
Culture Documents
1
• My office is located in 1001 Joyner library,
room 1006
• Email: bianh@ecu.edu
• Tel: 252-328-5428
• You can download sample data files from:
http://core.ecu.edu/ofe/StatisticsResearch/
2
• What is bivariate relationship?
–The relationship of two
measures/variables.
–The direction of the relationship
–The strength of the relationship
–Whether the relationship is statistically
significant
• Measures of association—a single
summarizing number that reflects the
strength of the relationship. This statistic
shows the magnitude and/or direction of a
relationship between variables.
• Magnitude—the closer to the absolute value
of 1, the stronger the association. If the
measure equals 0, there is no relationship
between the two variables.
• Direction—the sign on the measure
indicates if the relationship is positive or
negative. In a positive relationship, when
one variable is high, so is the other. In a
negative relationship, when one variable
is high, the other is low.
• Measurement
• Nominal: Numbers that are simply
used as identifiers or names represent
a nominal scale of measurement such
as female vs. male.
• Ordinal: An ordinal scale of measurement
represents an ordered series of relationships
or rank order. Likert-type scales (such as "On a
scale of 1 to 10, with one being no pain and
ten being high pain, how much pain are you in
today?") represent ordinal data.
• Interval: A scale that represents quantity and
has equal units but for which zero represents
simply an additional point of measurement is
an interval scale. The Fahrenheit scale is a
clear example of the interval scale of
measurement. Thus, 60 degree Fahrenheit or -
10 degrees Fahrenheit represent interval data.
• Ratio: The ratio scale of measurement is
similar to the interval scale in that it also
represents quantity and has equality of
units. However, this scale also has an
absolute zero (no numbers exist below
zero). For example, height and weight.
• Relationship based on measurement
–Two scale variables (two quantitative
variables)
–Two categorical variables (nominal or
ordinal)
–One categorical and one scale variable
• Statistical analyses for testing
relationships between two
quantitative variables
–Correlation
–Linear regression
• Pearson Correlation r
–Measures linear association
–It is the standardized regression
coefficient (doesn’t depend on
variables’ units)
–It is between -1 and +1
–The larger the absolute value of r, the
strongly the degree of linear
association
• Pearson Correlation r
–Pearson’s correlation coefficient
assumes that each pair of variables is
bivariate normal.
–There is no causal relationship
between two variables.
• Example: we want to know the
correlation between height and weight.
–Step 1: check the linear relationship
between two variables using Scatter
plot.
–Step 2: use Bivariate correlation to get
Pearson correlation.
• In SPSS, got to Graphs > Chart Builder
• Pearson correlation r: go to Analyze >
Correlate > Bivariate
• SPSS output
• Use Linear regression to test the
relationship between weight and height.
–We want to use height to predict
weight .
• Regression Equation
– Ypredicted = a + bx
Y
Regression line
Intercept
0
Slope X
• Residuals
–The difference between observed and
predicted values of the dependent
variable.
–We use residuals to check the
goodness of the prediction equation.
• Least square
– We use Least Square Criterion to estimate parameters.
– Lease Square means the sum of the squared estimated
errors of predictions is minimized.
48
• Logistic regression equation
– log(p/1-p) = a + bx
– Logit (p) = a + bx
p: probability of a case belonging to category
1
p/1-p: odds
a: constant
b: regression coefficient, about whether p
increases/decreases as x increases
• Odds
Odds of Success = Probability of
Success/Probability of failure
If Probability of Success = .75, then the
probability of failure = 1 - .75=.25, then
odds of Success = .75/.25 =3
• Odds
–Odds of Success > 1 means a success is
more likely than a failure.
–Odds of Success < 1 means a success is
less likely than a failure
• Odds ratio is the ratio of odds
• Example: we want to know the
association between Ever feel sad or
hopeless (Q26r) (independent
variable) and marijuana use (Q49r)
(Dependent variable)
• Contingency table
• Q49r is our dependent variable, in another
word, we want to know the probability of
using marijuana (Use group and coded as 1).
• The probability of a student using
marijuana is 3998/13242=.302
• We want to know whether this
proportion is the same for all levels of
Sad group.
• Odds of a Sad student using marijuana =
1345/2653 = .51
• Odds of a Happy student using marijuana =
1999/7245 = .28
• Odds ratio of using marijuana between Sad
and Happy students = .51/.28 = 1.82 ( Happy
group is the reference category)
• Logistic regression
• Go to Analyze > Regression >Binary
Logistic
• Click Categorical: No group is the reference
Check First as
Reference
category and click
Change
• Click Options
• SPSS output
We only have one model here. That is why the results of step, block,
and model are same.
• SPSS output