HYPOTHESIS/SIGNIFICANCE TESTING

 A hypothesis is simply a statement, an assumption, or some testable belief or opinion.

 Hypothesis testing is the process by which the belief is tested by statistical means.
 The essence of testing statistical hypothesis is to decide whether the claim (research
hypothesis) is supported by empirical evidence obtained through a random sample.

Notations:
 A hypothesis is denoted by H.
 Ho=Null Hypothesis (a hypothesis of “no difference” or “no relationship”)
 H1 or Ha = Alternative Hypothesis (Research Hypothesis)

Steps:
 Formulate the null and alternative hypotheses
 Establish the significance level and identify the acceptance and rejection
 Select the test statistic and procedure
 Collect the sample and compute the value of the test statistic
 Make the decision (accept or reject the null hypothesis).

Parametric & Non Parametric Tests

Parametric tests:
To use parametric tests, the following assumptions should hold:
 The observations must be independent i.e. the selection of any one case should not affect the
chances for any other case to be included in the sample
 Its observations should be drawn from normally distributed populations
 These populations should have equal variances
 The measurement scales should be at least interval so that arithmetic operations can be used
with them

Note:
 Parametric tests are more powerful than non-parametric tests because their data are derived from
interval and ratio measurement.
 They are more useful the more you know about your subject matter, since knowledge about
subject matter can be built into parametric models.

Non- parametric tests:

 Also known as distribution free tests or flexible functional forms
 One must use statistical technique called non-parametric if it satisfies at least one of the
following five types of criteria.
 The data entering the analysis are enumerative (i.e. count data representing the number of
observations in each category or cross-category)
 The data are measured and/or analyzed using a nominal scale of measurement
 The data are measured and/or analyzed using an ordinal scale of measurement
 The inference does not concern a parameter in the population distribution
 The probability distribution of the statistic upon which the analysis is based is not dependent
upon specific information or assumption about the population(s) which the sample(s) are

1
 drawn but only general assumptions, such as a continuous and/or symmetric population
distribution.

One and two sided tests

 One - sided (tailed) test

This is a statistical test in which the alternative hypothesis specifies that the population parameter
lies entirely above, or entirely below, the value specified in the null hypothesis.

e.g. Ho:  = 50,000

Ha:  > 50,000 is a one-sided test since Ha specifies that  lies on one particular side of
50,000

 Two- sided (tailed) test

An alternative hypothesis that specifies that the parameter can lie on either side of the value
indicated by Ho is called a two-sided or two-tailed test.

e.g. Ho:  = 50,000

Ha:   50,000 is a two-sided test

Type I & Type II Errors

 Type I error
Committed by rejecting a true null hypothesis (significance level)

 Type II error
Committed by failing to reject a false null hypothesis

SELECTING A TEST

Consider the following questions:

 Does the test involve one sample, two samples, or K (K>2) samples?
 If two samples or K samples are involved, are the individual cases independent or related?
 Is the measurement scale nominal, ordinal, interval, or ratio?
 What is the sample size? Is there several samples, are they equal in size?

PARAMETRIC TESTS

1. One-sample tests
 Used when we have a single sample and wish to test the hypothesis that it comes from a specific
population
 Following questions are encountered:
 Is there a difference between observed frequencies and the frequencies we would expect,
based on some theory?
 Is there a difference between observed and expected proportions?

2
  Is it reasonable to conclude that a sample is drawn from a population with some specified
distribution? (Normal, Poisson, etc)?
 Is there a significant difference between some measures of central tendency ( X ) and its
population parameter (µ)?

 Test for sample mean

Hypotheses
H 0 :   0
H1 :    0 OR H1 :    0
X  0
Test Statistic : Z 
/ n
when n  30
X  0
Test Statistic: t 
S/ n
when n  30, df  n  1

 Test for sample proportion

Hypotheses
H0 :   0
H1 :    0 OR H1 :    0
P 0
Test Statistic : Z 
 0 (1   0 )
n
when n  30
P 0
Test Statistic: t 
 0 (1   0 )
n
when n  30, df  n  1

2. Two independent samples

 Test for means

Hypotheses and alternatives

H 0 : 1   2
H1 : 1   2 (two  tailed test)

3
H 0 : 1   2
H1 : 1   2 (one  tailed test)

Test statistic(l arg e samples) Test statisticsmall samples

X1  X 2 X1  X 2
Z t
 12  22 S12 S22
 
n1 n2 n1 n2

Or when  1   2 ,
X1  X 2 S12 (n1  1)  S 22 (n2  1)
t Where, S p2 
S p2 S p2 n1  n2  2

n1 n2

 Test for proportions

Hypotheses and alternatives

H0 : 1   2
H1 :  1   2 (two  tailed test)

H0 :1   2
H1 :  1   2 (one  tailed test)

Test statistic(l arg e samples) Test statistic( small samples)

P1  P2 P1  P2
Z t
P1 (1  P1 ) P2 (1  P2 ) P1 (1  P1 ) P2 (1  P2 )
 
n1 n2 n1 n2

3. Two related sample tests

 These concern those situations in which persons, objects, or events are closely matched or the
phenomena are measured twice.
 In this case, paired t-test is appropriate.

Null hypothesis and alternatives

H 0 : d  0
H1 :  d  0 Or H1 :  d  0

Test statistic

4
 d
t
Sd / n

Where

S 2

 (d  d ) , 2

d
d , & d  X Y
n 1
d
n
NON-PARAMETRIC TESTS
Chi-square (   tests )
2

 Sign-tests (one and two sample tests)
 Mann-Whitney tests – to be used with two independent groups (analogous to the independent
groups t-test)
 Wilcoxon Matched pair test (to be used with two related groups (i.e. matched or repeated
groups (analogous to the related samples t-test
 Kruskall-Wallis – to be used with two or more independent groups (analogous to the one-way
ANOVA)
CHI SQUARE APPLICATIONS

Chi-square Test
 This is probably the most widely used non-parametric test of significance
 It is particularly useful in tests involving nominal data, but can be used for higher scales.
 It is useful in cases of one-sample analysis, two independent samples or k-independent
samples
 It must be calculated with actual counts rather than percentages.
Characteristics of the Chi-Square Distribution
 It is positively skewed
 It is non-negative
 It is based on degrees of freedom
 When the degrees of freedom change, a new distribution is created
Test of independence (Contingency Table Analysis)
 If we classify a population into several categories represented with two (or more attributes) we
can use a chi-square test to determine whether the two attributes are independent of each other.
 Contingency table analysis is used to test whether two traits or variables are related
 Each observation is classified according to two variables
Hypotheses:
Ho : The two classifications are independent (i.e. no relation between classes)
H1:The classifications are not independent;

Where O = Observed frequency

(O  E) 2
χ2  
E
5
 E = Expected frequency

 The degrees of freedom is equal to: (Number of Rows-1)(Number of Columns-1)

 The expected frequency is computed as follows:
(Row total)(Column total)
E
(Grand total)

 The 2 value obtained from the formula is compared with the value from the 2 table for a
given significance level and the number of degrees of freedom.

Goodness-of-Fit Test
In addition to the use of 2 for comparing observed and expected frequencies, the 2 test can be
used to determine how well empirical distribution (i.e. those obtained from a sample data) fit
theoretical distribution such as the normal, Poisson, and Binomial. That is, chi-square test of
"goodness of fit” tests the appropriateness of a distribution.

1. Equal Expected Frequencies

Let f 0 and f e be the observed and expected frequencies respectively.

Hypotheses:

H 0 : No difference between
H1 : There is a difference between
The test statistic is:  2   (f0  f e )
2

fe
The critical value is a chi-square value with (k-1) degrees of freedom, where k is the number of
categories

2. Unequal Expected Frequencies

The purpose is to test whether the observed frequencies in a frequency distribution match the
theoretical distribution. The expected frequencies are calculated in accordance with the assumed
distribution characteristics based on appropriate tables (Normal, Poisson, etc) and 2 calculated in
the manner previously described.