THE SIX-STEP MODEL FOR HYPOTHESIS

TESTING
Step 1. Stating the null hypothesis (H o) and its alternative
hypothesis (H1)
Step 2. Choosing a statistical test
Step 3. Selecting the level of significance and the sample size (N).
Step 4. Selecting the sampling distribution and establishing the
critical
Region
Step 5. Collecting data and computing the statistical test
Step 6. Making a decision
Step 1.
Stating the null hypothesis (Ho) and
alternative hypothesis (H1)
In conducting a research with statistical tool, we usually
express the hypothesis in its null form as null hypothesis
(Ho) initially and then express it in an alternative hypothesis
(H1).

A null hypothesis is a statistical hypothesis which states that

the observed is most likely due to chance variation while an
alternative hypothesis is one which states that the observed
outcome is due to factors other than chance variation.
x1122 222 0
 2 2
The form of the null hypothesis varies depending on the
test being conducted. For a single mean, the null
hypothesis would state that there is no significant
difference between the sample mean and the population
mean.
H o : X 
The null hypothesis is the central element in any test of
hypothesis because the whole test is aimed at rejecting or
accepting the null hypothesis.
An alternative hypothesis is the operational statement of
the researcher’s prediction while the research hypothesis
is a statement that is based on the information gathered
from the random sample(s) or from the data being tested.
One form of the research hypothesis might be

H o : X 
Step 2.
Choosing a statistical test
Step 3.
Selecting the level of significance and
the sample size (N)
Step 4.
Selecting the sampling distribution
and establishing the critical region
Step 5.
Collecting data and computing the
statistical test
Step 6.
Making a decision
HYPOTHESIS TESTING
USING THE SPSS
t test
• one sample
• independent
samples
• paired samples
t test: one sample

The one-sample t-test is used

to determine whether a sample
comes from a population with a
specific mean.
A car company claims that their Super Spiffy Sedan averages 31
mpg. You randomly select 8 Super Spiffies from local car
dealerships and test their gas mileage under similar conditions.
You get the following MPG scores:
30, 28, 32, 26, 33, 25, 28, 30

Does the actual gas mileage for these cars deviate

significantly from 31
(alpha = .05)?
1. Click Analyze > Compare Means > One-
Sample T Test... on the main menu:
2. You will be presented with the One-Sample T
Test dialogue box,
3. Transfer the dependent variable into the ‘test
variables’ box and the population mean you
are comparing the sample against into the
‘test value’ box
4. Click on options to change the confidence
interval
5. Click on continue
t test: independent samples

The Independent Samples t Test compares the

means of two independent groups in order to
determine whether there is statistical evidence
that the associated population means are
significantly different. The Independent
Samples t Test is a parametric test.
est: independent samples
est: independent samples
 Twenty participants were given a list of 20 words to process. The 20
participants were randomly assigned to one of two treatment conditions.
Half were instructed to count the number of vowels in each word (shallow
processing). Half were instructed to judge whether the object described by
each word would be useful if one were stranded on a desert island (deep
processing). After a brief distractor task, all subjects were given a surprise
free recall task. The number of words correctly recalled was recorded for
each subject. Here are the data:
Shallow processing: 13, 12,11,9,11,13,14,14,14,15
Deep Processing: 12,15,14,14,13,12,15,14,16,17

Did the instructions given to the participants

significantly affect their level of recall (alpha = .05)?
1. click variable view, values, add value ‘1‘ and type
appropriate label. click add.
2. Go to data view. Click Analyze > Compare Means >
Independent-Samples T Test... on the top menu,
3. You will be presented with the Independent-Samples T
Test dialogue box
4. Transfer the dependent variable into the ‘test variable’
box and the independent variable into the ‘grouping
variable’ box
5. click on define groups
6. Enter the specified values for each group as you have
typed in step 1.
7. click continue
8. if you need to change the confidence level limits click on
test: paired samples
The dependent t-test (called the paired-samples t-test in
SPSS Statistics) compares the means between two related
groups on the same continuous, dependent variable.

For example, you could use a dependent t-test to

understand whether there was a difference in smokers'
daily cigarette consumption before and after a 6 week
hypnotherapy programme (i.e., your dependent variable
would be "daily cigarette consumption", and your two
related groups would be the cigarette consumption values
"before" and "after" the hypnotherapy programme).
test: paired samples
HO:
There is no significant
difference in the blood
pressure of patients
before and after taking
oral contraceptives.
tests of correlation
• Pearson product moment correlation
• Phi-coefficient
• Point biserial
• Spearman’s rho
• Rank biserial
• Biserial correlation
• tetrachoric coefficient
• partial correlation
Pearson-product moment correlation
This determines the strength of the linear relationship or association
between two variables that have been measured at the interval or ratio
level.

1. click Analyze > Correlate > Bivariate.

2. select the variables in the list on the left and click the
blue arrow button to move them to the right, in
the Variables field.
3. Clicking Options will open a window where you can
specify which Statistics to include
 HO:
The blood glucose
level of patients is
not related to age.
To test if the relationship is
significant:

N 2
t r 2
1 r
Phi- coefficient

This technique is used to test the extent of association

between two sets of attributes measured on nominal scales,
each of which can take on only two values. This means that
the variables are dichotomous in nature.
 HO:
The type of music genre that the subjects listened
to is not significantly related to their state of stress.
To test for the significance of the correlation
Point biserial

This is a measure of association between one

variable measured on a real dichotomous nominal
scale and the other variable measured in the
interval scale.
 POINT BISERIAL CORRELATION
EXAMPLE
• A RESEARCHER WISHES TO DETERMINE IF A SIGNIFICANT
RELATIONSHIP EXISTS BETWEEN THE GENDER OF THE
WORKER AND IF THEY EXPERIENCE PAIN WHILE
PERFORMING AN ELECTRONICS ASSEMBLY TASK.
• THE INDEPENDENT VARIABLE IS THE QUESTION WHICH
ASKS “WHAT IS YOUR GENDER?___ MALE ____ FEMALE”
(DICHOTOMOUS)
• THE DEPENDENT VARIABLE IS FROM THE QUESTION THAT
ASKS “HOW MANY YEARS HAVE YOU BEEN PERFORMING
THE TASKS? ____ YEARS” (RATIO)
 Attendance in Score in
Pupil Nursery Communication
Skills Test
A 1 30
B 1 27
C 0 19
D 0 15
E 1 26
F 0 22
G 1 23
H 1 21
I 0 16
J 1 17
K 0 18
L 1 28
1. click Analyze > Correlate > Bivariate.
2. select the variables in the list on the left and click the
blue arrow button to move them to the right, in
the Variables field.
3. Clicking Options will open a window where you can
specify which Statistics to include
To determine the significance of the association between
the two variables, use t-test.
Spearman rho

This is a measure of association between two variables,

which are both measured at least in the “continuous”
ordinal scale; that is, the individuals under study can be
ranked in two ordinal series.
A science teacher hypothesized that there is an association between interest in
mathematics and problem solving ability in physics.
Problem Solving Ability
Interest in Math
in Physics
Student x Rank y Rank
A 23 1 92 1
B 21 2 85 2.5
C 20 3 83 5
D 19 4 79 7
E 18 5 84 4
F 15 6.5 75 8
G 15 6.5 85 2.5
H 14 8 82 6
I 13 9 70 9
J 12 10 68 11
K 10 11 69 10
L 9 12 67 12
1. click Analyze > Correlate > Bivariate.
2. select the variables in the list on the left and click the
blue arrow button to move them to the right, in
the Variables field.
3. Clicking Options will open a window where you can
specify which Statistics to include
To determine the significance of the association between
the two variables, use t-test.

N 2
t rs 2
1  rs
Rank biserial correlation

This determines the strength of the linear relationship

between one variable measured on a dichotomous normal
scale and the other variable measured in an ordinal scale.
 Ownership Interest
of
Textbook In
Student (y)
Does having Learning
Statistics
a textbook in (x)
Statistics A 1 18
affect the B 0 16
C 0 11
interest in
D 1 20
learning E 1 14
Statistics? F 1 15
G 0 10
H 0 8
I 1 12
J 1 13
In SPSS:

•Click Analyze → Correlate → Bivariate.

•Add your variables, deselect Pearson

(the default) and click Spearman.*

•Click OK.
To determine the significance of the association between
the two variables, use t-test.

r N 2
t
2
1 r
 Tetrachoric Coefficient

Tetrachoric correlation is used to measure

rater agreement for binary data; Binary data
is data with two possible answers — usually
right or wrong.

SPSS Statistics currently does not have a procedure that

will produce tetrachoric or polychoric correlations.
Partial Correlation Coefficient

This can be used when a researcher would

like to observe how a bivariate relationship
exists having controlled the influence of a
third variable.
 English
Math Skills Test Science IV
Proficiency
Student Score Achievement
Score
A 80 85 84
B 87 84 85
C 82 80 81
D 91 90 92
E 85 86 88
F 93 91 94
G 81 79 82
H 88 88 86
I 86 83 87
J 94 95 93
K 83 82 84
L 92 94 91
M 89 87 89
N 90 89 92
O 95 96 95
1. Click Analyze > Correlate > Partial... on the
menu system,
2. Transfer the variables into the corresponding
boxes
3. click the options box
4. Tick the Means and standard
deviations and Zero-order correlations checkbox
in the –Statistics– area,
5. Click continue
6. Click OK
Nonparametric tests are used:
(a) when the sample is so small that the sampling
distributions of the statistics don’t assume a normal
distribution,
(b) when no assumption can be made about the shape of
the distribution from which the sample is drawn, and

(c) when categorical data (nominal or ordinal) are

used.
The Chi- square Goodness of Fit Test
The chi-square goodness-of-fit test is a single-sample
nonparametric test, also referred to as the one-sample
goodness-of-fit test or Pearson's chi-square goodness-of-fit
test.

It is used to determine whether the distribution of cases

(e.g., participants) in a single categorical variable (e.g.,
"gender", consisting of two groups: "males" and
"females") follows a known or hypothesised distribution
(e.g., a distribution that is "known", such as the proportion
of males and females in a country; or a distribution that is
"hypothesised", such as the proportion of males versus
A sample of 240 coffee drinkers was asked to choose
among the six brands of coffee. The results of the survey
are as follows:
1. Click Analyze > Nonparametric
Tests > Legacy Dialogs > Chi-square...
2. Transfer the variables into the box
3. Keep the All categories equal option selected
in the –Expected Values– area as we are
assuming equal proportions for each
category.
df = k – 1,
The Runs test for Randomness

a non-parametric method that is used in

cases when the parametric test is not in
use. In this test, two different random
samples from different populations with
different continuous cumulative distribution
functions are obtained.
Conduct a runs test at  = .05 on the following data
series:

F F F F M M M M M F F F F M M M F
F F F M F F F F
Nemar Test for Significance of Changes
particularly applicable to “before and after” design
in which each subject is used as its own control
and measurements are made before and after the
treatment.

It can be considered to be similar to the

paired-samples t-test, but for a
dichotomous rather than a continuous
dependent variable.
 For example, you could use the McNemar
test to determine whether the proportion of
participants who had low self-esteem (as
opposed to high self-esteem) before a series
of counselling sessions (i.e., an intervention)
decreased after the intervention (i.e., your
dependent variable would be "level of self-
esteem", which has two categories: "low"
and "high").
 A researcher wanted to investigate the impact
of an intervention on smoking. In this
hypothetical study, 50 participants were
recruited to take part, consisting of 25 smokers
and 25 non-smokers. All participants watched
an emotive video showing the impact that
deaths from smoking-related cancers had on
families. Two weeks after this video
intervention, the same participants were asked
whether they remained smokers or non-
 Before After Frequency
Non-Smoker Non-Smoker 20
Non-Smoker Smoker 5
Smoker Non-Smoker 16
Smoker Smoker 9
1.For SPSS Statistics versions 18 to 22, click Analyze
> Nonparametric Tests > Legacy Dialogs > 2
Related Samples... on the main menu (as shown
below) but, for older versions of SPSS Statistics,
click Analyze > Nonparametric Tests > 2 Related
Samples... on the main menu.

2.Transfer the variables to the corresponding boxes

If the statistical significance level (i.e., p-value) is less than
.05 (i.e., p < .05), you have a statistically significant result
and the proportion of non-smokers before and after the
intervention is statistically significantly different.
Alternatively, if p > .05, you do not have a statistically
significant result and the proportion of non-smokers before
and after the intervention is not statistically significantly
The chi-square test of association

used to test hypothesis about frequencies obtained from two

independent samples.
 Is there a relationship between scientific literacy and class
standing? Are upperclassmen significantly different from the
underclassmen on this variable?
Scientific Class Standing Combined
Literacy
Below Above
Average Average
High 10 33 43
Average 30 40 70
Low 54 30 84
Total 94 103 197
1.Click Analyze > Descriptives Statistics
> Crosstabs... on the top menu,

2.Transfer the variables into the box

3.Click on statistics button. select chi square. click

continue

4.click cells. click observes, row, column, total

5. click continue
 Mann-Whitney U-Test

When variables are measured in ordinal scale, the Mann-

Whitney U-test may be used to test whether these two
groups come from the same population. This is one of the
most powerful of the nonparametric tests.
Degree of Anxiety of Female Students Degree of Anxiety of Male Students
Scores Rank Scores Rank
48 1 47 2
45 3 44 4
43 5 40 7
42 6 35 10
39 8 32 12
36 9 31 13
33 11 30 14
28 16 29 15
23 19 25 17
21 20 24 18
15 22 20 21
14 23 12 24
 Wx = 143  Wy = 157
1. Click Analyze > Nonparametric Tests
> Legacy Dialogs > 2 Independent
Samples... on the top menu,

2. Transfer the data into the box

3. Click define groups

4. click continue

5. select mann-whitney u test