Professional Documents
Culture Documents
Research Methods
William G. Zikmund
Chapter 21:
Univariate Statistics
introduction
Statistical analysis can be divided into several
groups
Univariate statistical analysis: tests of hypotheses
involving only one variable
Bivariate statistical analysis: tests of hypotheses
involving two variables
multivarate statistical analysis: tests of hypotheses
involving three or more variables or set of
variables
Hypothesis testing
Hypotheses are formal statements of
explanations stated in a testable form
Relational hypothesis: examine how change in
one variable vary with change in another
variable
Hypothesis about difference between groups:
examine how some variables varies from one
group to another
Hypothesis about difference from some
standard:
Hypothesis
Unproven proposition
Supposition that tentatively explains certain
facts or phenomena
Assumption about nature of the world
An unproven proposition or supposition that
tentatively explains certain facts or
phenomena
Null hypothesis
Alternative hypothesis
Null Hypothesis
Statement about the status quo
No difference exists between groups being
compare
No relationship between variable being
analyzed
Alternative Hypothesis
Statement that indicates the opposite of the
null hypothesis
There is significant difference between
groups being compared
There is a significant relationship between
variable
H o : 3.0
H 1 : 3 .0
Null is true
Null is false
Reject null
Correctno error
Type I
error
Type II
error
Correctno error
Decision
Accept Ho
Reject Ho
Ho is true
Ho is false
Correct--no error
Type II error
Type I error
Correct--no error
When the null hypothesis is true and you reject it, you make a type I error.
The probability of making a type I error is , which is the level of significan
you set for your hypothesis test. An of 0.05 indicates that you are willing
accept a 5% chance that you are wrong when you reject the null hypothesi
To lower this risk, you must use a lower value for . However, using a lowe
value for alpha means that you will be less likely to detect a true difference
if one really exists.
Significance Level
Critical probability in choosing between the null
hypothesis and the alternative hypothesis
The acceptable level of type one error
significance or alpha () level, which is also set
ahead of time, usually at 0.05 (5%)
Acceptable amount of error
Probability Level selected is typically .05 or .01
Too low to warrant support for the null
hypothesis
P-value
P-ValueThe probability value (p-value) of a statistical hypothesis
test is the probability of getting a value of the test statistic as
extreme as or more extreme than that observed by chance alone, if
the null hypothesis H0, is true.
It is the probability of wrongly rejecting the null hypothesis if it is
in fact true.
Scale of measurement
PARAMETRIC
STATISTICS
NONPARAMETRIC
STATISTICS
Parametric statistics
When data are interval or
ratio
Sample size is large
Numbers with known,
continuous distribution
Non-parametric statistics
Numbers do not confirm to
a known distribution
When data are either ordinal
or nominal
A Sampling Distribution
A Sampling Distribution
A Sampling Distribution
UPPER
LIMIT
LOWER
LIMIT
Critical values of
Critical value - upper limit
S
ZS X or Z
n
1.5
3.0 1.96
225
Critical values of
3.196
Critical values of
Critical value - lower limit
- ZS X or - Z
3.0 - 1.96
1 .5
225
S
n
Critical values of
2.804
Region of Rejection
LOWER
LIMIT
UPPER
LIMIT
Hypothesis Test
2.804
3.196
3.78
Calculating Zobs
x
z
sx
obs
Z obs
SX
Z obs
3.78
3.78 3.0
SX
.1
0.78
.1
7.8
t-Distribution
Degrees of Freedom
Abbreviated d.f.
Number of observations-Number of
constraints
X tc.l . S X
Upper limit X tc.l .
or
S
n
S
n
= population mean
X = sample mean
tc.l . = critical value of t at a specified confidence
SX
S
n
level
= standard error of the mean
= sample standard deviation
= sample size
X t cl s x
X 3.7
S 2.66
n 17
H 0 : 20
H1 : 20
SX S / n
5 / 25
1
Lower limit :
tc.l . S X 20 2.064 5 / 25
20 2.0641
17.936
Upper limit :
t c.l . S X 20 2.064 5 / 25
20 2.0641
20.064
tobs
X
22 20
SX
1
2
1
2
Chi-Square Test
(Oi Ei )
x
Ei
Chi-Square Test
x = chi-square statistics
Oi = observed frequency in the ith cell
Ei = expected frequency on the ith cell
Chi-Square Test
Estimation for Expected Number
for Each Cell
Eij
Ri C j
n
Chi-Square Test
Estimation for Expected Number
for Each Cell
Ri = total observed frequency in the ith row
Cj = total observed frequency in the jth column
n = sample size
O1 E1
E1
O2 E2
E2
60 50
50
40 50
50
H0 : . 5
H1 : . 5
Sp
0.6 0.4
100
.0024
.24
100
.04899
.6 .5
p
Zobs
.04899
Sp
.1
2.04
.04899
pq
n
Sp
(.2)(.8)
1200
Sp
.16
1200
Sp .000133
Sp . 0115
pq
n
Sp
(.2)(.8)
1200
Sp
.16
1200
Sp .000133
Sp . 0115
p
Sp
.20 .15
.0115
.05
Z
.0115
Z 4.348
The Z value exceeds 1.96, so the null hypothesis should be rejected at the .05 level.
Indeed it is significantt beyond the .001
Z