TESTING OF HYPOTHESIS

PRESTLINE JOSE (B2147)

  Inferential Statistics
 Hypothesis: A hypothesis is a proposition about the
status of events or about relation between variables
(Educated Guess)
 It has to be tested in the course of a research
 Null Hypothesis: No relationship between variables
INTRODUCTION Ex: There is no relationship between crop yield and use
of fertilizers
 Alternate Hypothesis: Existence of relationship
Ex: There is a relationship between crop yield and use of
fertilizers
  Type I error: Rejecting a null hypothesis when it is true
 Probability of Type I error is called level of significance
denoted by alpha
TYPES OF  Type II error: Accepting a null hypothesis when it is
false
ERROR  Probability of Type II error is denoted by Beta
 Power of the Test: Probability that we will correctly
reject the null hypothesis when it is false. It is given by
the formula : 1 – β
  Test statistic: The decision to accept or reject a null
hypothesis is made on the basis of a statistic computed from
the sample. Such a statistic is called the test statistic. The
commonly used test statistics are t, z, F etc.
 Sampling distribution: Select a number of random samples
IMPORTANT of definite size from a given population and calculate some
statistic, from each of these samples. Probability distribution
TERMS of that sample statistic is called sampling distribution.
 Standard Error: SD of a sampling distribution of a statistic is
called SE
 Degree of freedom: Number of independent observations
which is obtained by subtracting the number of constraints
from the total number of observations
 i. State the null hypothesis
ii. Choose the statistical test
STEPS IN iii. Select the desired level of significance
HYPOTHESIS iv. Compute the calculated difference value
TESTING v. Obtain the critical test value
vi. Interpret the result
  T-test: Applications
 To test the given population mean when the sample is small and
the population SD is not given
 To test the equality of two sample means when the samples are
PARAMETRIC small and population SD is unknown.
TESTS:  Assumptions
T-TEST  The parent population is normal
Z-TEST  Population SD is unknown
F-TEST  Equality of two population means: Samples are independent and
population variance is assumed to be equal and unknown
 Example: A stenographer claims that she can take dictations at the
rate of more than 120 words per minute. Of the tests given to her
she could perform an average of 135 words with SD of 40. Is her
claim valid?
  Applications:
 To test the given population mean when the sample is large or
when the population SD is unknown
 To test equality of two sample means when the samples are
large or when the population SD is known
 To test population proportion
Z-TEST  Assumptions:
 Population is normal
 Population is large
 Ex: It is claimed that a random sample of 100 tyres with mean
life of 15269 km is drawn from a population of tyres which has
a mean life of 15200 km and SD of 1248 km. Test the validity
of the claim.
  To test significant difference between variances of two
populations
 Assumptions:
 Samples are drawn at random
 Population follows normal distribution
 Population SD are equal but unknown
F-TEST  Ex: The following table shows output by two workers A
and B
 Can we conclude that both these workers are equally
stable?
A 40 30 38 41 38 35
B 39 38 41 33 32 39 40 34
  Analysis of Variance
 Technique which analyzes the variances of two or more
comparable series (or samples) for determining the
significance of differences in their Arithmetic means,
and for determining whether different samples under
study are drawn from same population or not, with the
help of F test.
ANOVA  Evolved with agricultural research
 One-Way ANOVA: Classification based on single
criterion
 Ex: Productivity and machines used
 Two-Way ANOVA: Classification based on two criteria
 Ex: Change in productivity depending on machines and
workers
  Variance between samples (Residual Variance)

TYPES OF  Variance within the sample

VARIANCES  Variance about the sample

TREATMENT
PLOT NO. I II III IV
1 42 48 68 80
2 50 66 52 94
3 62 68 76 78
4 34 78 64 82
5 52 70 70 66
Total 240 330 330 400
  Chi-Square Test:
NON-
 Sign Test
PARAMETRIC
 Wilcoxon Matched –Pairs Test
TESTS  Mann-Whitney-Wilcoxon U-test
  It describes the magnitude of difference between
theory and observation
 Applications:
 Useful for the test of goodness of fit: It helps to
ascertain how well theoretical data fit into actual
CHI-SQUARE observation
TEST  Ex: Tossing of coins
 Useful for the test of independence of attributes:
association between two attributes
 Ex: Economic status of respondents and their attitude
towards elections
  Useful for testing homogeneity: Association between
different samples
APPLICATION  Ex: Attitude towards election among different villages

S OF CHI-  Useful for testing given population variance: Testing

whether the given population variance is acceptable or
SQAURE not on the basis of samples drawn from that population
 Ex: Variance of height among students of MBA B and
Variance of height of all Rajagirians
THANK YOU!