UNIT VI.

DESIGNS AND MEANING

 Research design makes researcher as efficient as
possible yielding maximal information with minimal
expenditure of efforts, time and money.
 Just as for better, economical and attractive

construction of a house we need a blueprint (map of

the house prepared by an expert architect.
 We need a research design or a plan in advance of

data collection and analysis for our research project.

 A good design minimises bias and maximises the

reliability of the data collected and analysed.

DFFIERENT RESEARCH DESIGNS
1. Descriptive Design
2. Correlation Design
3. Developmental Research Design
4. Epidemiological Design
5. Survey research Design
 EXPERIMENTAL DESIGNS

 The experimental design or the design of experiment

(DOE) is defined as the design of an information-
gathering experiment in which a variation is
present or not, and it should be performed under
the full control of the researcher. This term is
generally used for controlled experiments.
 COMPARISION IN PAIRS
 Pair wise comparison generally is any process of
comparing entities in pairs to judge which of each
entity is preferred, or has a greater amount of some
quantitative property, or whether or not the two
entities are identical.
 A paired comparison scale presents the respondent

with two choices and calls for a preference.

 Ex. The respondent is asked which colour he or she

likes better, red or blue and a similar process is

repeated throughout the scale items
 The method of paired comparisons is a “a simple and
direct way of collecting judgement data.
 RANDOMIZED BLOCK DESIGN
 A randomized block design is a type of experiment
where participants who share certain characteristics
are grouped together to form blocks, and then the
treatment (or intervention) gets randomly assigned
within each block.
 Ex. Two groups of patients each receiving specific

drug can be chosen with randomization..

PATIENT 1 2 3 5 8 11 14 16 18 20
NUMBER

DRUG 1 D1 D1 D1 D1 D1 D1 D1 D1 D1 D1
(ONE
WEEK)

PATIENT 4 6 7 9 10 12 13 15 17 19
NUMBER

DRUG 2 D2 D2 D2 D2 D2 D2 D2 D2 D2 D2
(ONE
WEEK)
 LATIN SQUARE DESIGN
 The Latin square design gets its name from the fact that
we can write it as a square with Latin letters correspond
to the treatments.
 A Latin square design is a method of placing

treatments so that they appear in a balanced fashion

within a square block or field. Treatments appear once
in each row and column.
 A Latin square is a block design with the arrangement of

V Latin letters into a V x V array (a table with V rows

and V columns). LSDs are often used in experiments
where subjects are allocated treatments over a giver
time period where time is thought to have a major
effect on the experimental response.
 LSD can have a maximum of only 16 plots
 A, B, C, D are types of treatments

 Each treatments appears only once in row (river

gradient, A) and column (road gradient, B)

 LSD is an experimental design frequently used in
agriculture research
 For example an experiment has to be made through

which the effects of five different fertilizers on the

yield on certain crop (i.e.) wheat.
 The varying fertility of the soil in different blocks in

which experiment is going to be performed otherwise

the result obtained may not be dependable because
the output effects not only fertilizers but also fertility
of the soil.
A B C D E

B C D E A

C D E A B

D E A B C

E A B C D
 PARAMETRIC TESTS NON PARAMETRIC TESTS

1. These tests are based on 1. NPT are not based on

assumptions about the assumptions, requires no
distribution of population assumptions.
from which the samples are 2. No assumptions are made
taken. regarding population.
2. Specific assumptions are 3. Doesn’t require the
made regarding the previous knowledge or
population. information about the
3. Requires previous population.
knowledge or information 4. NPT analysis to test group
about the population. median.
4. Parametric analysis to test 5. Samples are not
group means. necessarily independent.
5. Samples are independent. 6. Qualitative data.
6. Quantitative data. 7. Applicable for both
7. Applicable only for variable. variable and attributes.
8. Uses a normal probabilistic 8. The distribution is
distribution. arbitrary.
1. T-Test 1. Chi-square test
2. Paired T-Test 2. Sign test
3. ANOVA 3. Median test
4. MANOVA 4. Mann-Whitney
5. ANCOVA Test
5. Wilcoxon signed
Rank Test
6. Kruskal-Wallis
Test
7. Friedanan’s
ANOVA
Fisher's Exact Test
 Fisher's exact test is a statistical test used to determine if

there are nonrandom associations between two

categorical variables.
 Let there exist two such variables and , with and observed

states, respectively. Now form an matrix in which the

entries represent the number of observations in which and .
Calculate the row and column sums and , respectively, and
the total sum
 (1)

 of the matrix. Then calculate the conditional probability of

getting the actual matrix given the particular row and column
sums, given by
 (2)

 which is a multivariate generalization of the hyper-geometric

probability function. Now find all possible matrices of

non-negative integers consistent with the row and column
sums and . For each one, calculate the associated
conditional probability using (2), where the sum of these
 To compute the P-value of the test, the tables must then be
ordered by some criterion that measures dependence, and
those tables that represent equal or greater deviation from
independence than the observed table are the ones whose
probabilities are added together. There are a variety of
criteria that can be used to measure dependence. In
the case, which is the one Fisher looked at when he
developed the exact test, either the Pearson chi-square or the
difference in proportions (which are equivalent) is typically
used. Other measures of association, such as the likelihood-
ratio-test, -squared, or any of the other measures typically
used for association in contingency tables, can also be used.
 The test is most commonly applied to matrices, and is

computationally unwieldy for large or . For tables larger

than , the difference in proportion can no longer be used, but
the other measures mentioned above remain applicable (and
in practice, the Pearson statistic is most often used to order
the tables). In the case of the matrix, the P-value of the test
can be simply computed by the sum of all -values which are .
 For an example application of the test, let be a
journal, say either Mathematics Magazine or Science,
and let be the number of articles on the topics of
mathematics and biology appearing in a given issue
of one of these journals. If Mathematics Magazine has
five articles on math and one on biology,
and Science has none on math and four on biology,
then the relevant matrix would be
UNIT VII. TESTING HYPOTHESIS OR LEVELS
OF SIGNIFICANCE
A statistical test used in the case of non parametric independent
variables is called NPT.
NPT is defined as the hypothesis test which is not based on
underlying assumptions i.e. it does not require population’s
distribution to be denoted by specific parameters
Advantages of NPT
 Probability statement obtained from most of the
parametric tests are exact probabilities.
 If the sample are of size at small as to 6. There is

alternative of using NPT.

 These tests are usually easier to learn and apply them

to parametric tests
 These are suitable tests for treating observation from

samples drawn from several different population

 Applicable for both variable and attributes.
 DISADVANTAGES OF NPT
 There are no non parametric methods for testing
interaction in the analysis of variants.
 All the assumptions of parametric are met in the

data itself.
 tables of many non parametric tests are not

available readily with the researcher.

 UNIT IX: APPLICATION OF STATISTICS IN HEALTH
 Vital & Health Statistics
 is accumulated data gathered on live births, deaths,

migration, foetal deaths, marriages & divorces.

 The most common way of collecting information on

these events is through civil registration, an

administrative system used by governments to record
vital events which occur in their population
 Hospital vital statistics efforts to improve the quality

of vital statistics & people's health

 Vital statistics is a tool in developing public health
policy and initiatives.
 Health officials, health care professionals & scientific

community reply on accurate & timely data to deal

with out breaks of infectious diseases such as Ebola &
Covid-19.
 For describing vital events occurring in community we

need vital and health statistics.

 Vital events are those pertaining to human life such as

births, deaths, sickness, marriages, divorces,

migration etc.
 Vital statistics are generally obtained through the
sources of population census, sample surveys & vital
statistics registers.
 Vital statistics include the counts of births, deaths,

illness, movements & the various rates and ratios that

may be computed from them & utilised.
 The information obtained from hospital in door &
outdoor facilities regarding quality of care utilization
of services, quality of services delivered & other
hospital related administrative & logistic affairs is
called Hospital Statistics.
 Measures of evaluation of quality of care

 Helps in planning

 Allocation of resources in different areas

 Identify deficiencies at various level i.e. inpatient

process & out come of services.

 Evaluate effectiveness & efficiency of the

administration.
 TYPES OF HOSPITAL STATISTICS
 Daily census
 Daily average attendance

 Bed occupancy rate

 Bed turnover rate

 Total patients days care

 Vacancy rate

 Hospital beds

 Sanctioned beds

 Bed supply rate

 RATIOS
 Ratio is a result from dividing one quantity by
another. Mostly used when numerator & the
denominator are two distinct quantities. Ratio is used
in comparing the frequencies of two numerically
exclusive classes.
 Sex ratio= SR= F x 100

M
 Example: Sex ratio, Foetal death ratio, person doctor

ratio, person nurse ratio, person hospital bed ratio.

 From female percentage: 60% females. 60 /100 =
100 / (x+ 100) x= 66. Therefore, 66:100is the
conventional sex ratio. From male percentage: 60%
males. (100 -60) / 100 = 100 / (x+ 100) x= 150.
Therefore, 150:100is the conventional sex ratio.
Percentages.
 RATES
 Proportion between total events & population over a
period of time for a given geographical area.
 Example: Maternal mortality rate, infant mortality

rate, hospital death rate

 RATES
Proportion between total events & population over a
period of time for a given geographical area.

No. o events which occur among

specific group of population of a given
geographical area during a given
time
Specific rate means = _________________________________
x 1000
Mid year population of the
specific group of
population in the same
geographic area during
a given area
CRUDE DEATH RATE – CDR

Annual Death (D)

CDR = ___________________ x 1000
Annual Mean Population (P)
 CALCULATE THE CDR FOR THE FOLLOWING
Age group(years) No.of persons(1000) No. of
deaths

Bellow 10 12 150
10 – 30 20 110
30 – 45 35 380
46- 70 24 210
Above 70 15 540
Total 106 Total 1390
CDR = D/P x1000
= 1390/106 x1000
= 1390/106
= 13.11 per thousand
 SPECIFIC DEATH RATE (SDR)
No. of deaths in the specific age group
SDR = _______________________________ x
1000
Population in that age group
Find SDR of the following

Age group(Years) Population(1000) No. of deaths SDR

0 - 15 6 150 150 x1000/6x1000 =25
15 – 40 20 180
180x1000/20x1000= 9
40 – 60 10 120
120x1000/10x1000=12
Over 60 4 160 160x1000/4x1000=
40
 STANDARDIZED DEATH RATE (STDR)
EMi Spi
STDR = ______
ESPi
i = age group
Mi = SDR of its age group
Spi = Standard population of its age group

Note: Whenever the population under study is taken as

standard population then STDR = CDR
 TRENDS
 Trend studies – A general direction in which a
situation is changing or developing. These studies
help to investigate a sample from a general
population over a time with respect to some
phenomenon. Trend studies permit researchers to
examine pattern and rate of changes & to make
prediction about future direction based on previous
identified patterns and rates of changes.
BIRTH & DEATH RATES
FERTILITY, MORBIDITY &
MORTALITY
 UNIT X: USE OF STATISTICAL PACKAGE
 SPSS Statistical Package for Social Sciences or some

times it is also called as Social Sciences

 Where you will get this package

 It is a software example windows 7, 8, , 11,,

 The techniques we learnt before Chi-Sqare, r,

p,z,,,,can be studied and calculated SPSS

 It is only one software

 It was developed in the

 IBM software company purchased SPSS software so it

called as IBM-SPSS
 SPSS-1,2,3, , ,recent version 23
 This software is sold by IBM and it is very costly like

Rs.10 lakhs, 5 lakhs.

 Individually we cont buy usually the research

institutions will buy it

 USE OF SPSS SOFTWARE
 To do descriptive statistics, we can create tables, pie
charts, diagrams.
 we can find measures of Mean, Median, Mode ,

Standard deviation ,,,,,,,,

 We can do inferential statistics also example test of

significance , Chi-Sqare, t-test, correlations r and z

tests.
 Reliability of scale tool construction, validity and

reliability
 Consistency of instrument/tool any body can do we

will get the same result

 USE OF COMPUTERS IN RESEARCH
 Performing calculation at the speed of light
 Data analysis occurring large research project

 Huge storage of data

 Faster retrieval

 Added quality to research activity

 Trial and error process frequently employed

 Using computers and the internet, it is possible to

access a wider library of information at a faster rate.

 The computer has made the whole research process
easier
 It has allowed researcher all over the world to come

together and refine their work. Thus improving the

entire psychological field.
 The development of research today is actually tied in

with computers. They are not just used to transfer the

information but to actually conduct the research.
 Data organisation and coding

 Storing the data

 Selection of appropriate statistical

measures/techniques
 Selection of appropriate software package
 Execution of the computer program

 Problem identification

 Literature search

 Research design

 Data collection and analysis

 Research dissemination