Professional Documents
Culture Documents
BBT BFT-676 BHT 656 Rm-U 3
BBT BFT-676 BHT 656 Rm-U 3
2
Processing and Analysis of Data
Data Analysis is process, the technique through which meaning, is attached to the
responses given by the respondents/subjects.
3
Processing and Analysis of Data
The data, after collection, has to be processed and analysed in accordance with the
outline laid down for the purpose at the time of developing the research plan. This is
essential for a scientific study and for ensuring that we have all relevant data for
making contemplated comparisons and analysis. Technically speaking, processing
implies editing, coding, classification and tabulation of collected
data so that they are amenable to analysis.
Common perception about data analysis: Only Quantitative data is analyzed
Reality: Data analysis is required for both Qualitative and Quantitative research
The technique of analysis may be different but all kind of information (Data) needs
to be analyzed for giving meaning to those information
4
Processing and Analysis of Data
5. Tabulation
6. Data presentation
7. Data Cleaning
8. Data Adjusting
6
Why Data Analysis?
In Quantitative Research data analysis is done to figure out the range of responses.
The data analysis enables to translate the responses into numbers which is essential
in any quantitative research.
In Qualitative Research data analysis is done to give meaning to the data which is
gathered using different research methods. In Qualitative Research in data analysis
the focus is more on attaching meaning to the information.
7
Analysis of Data in Qualitative Research
In Qualitative Research data is collected by using Case Study Method, Observation Method,
Qualitative Interview method, Focus Group Interviews and similar such qualitative methods.
The analysis of Qualitative data is not expressed in numerical form.
In Qualitative research, the data is analyzed in terms of stated experience of the respondents.
(statement, experience etc.)
The data is analyzed to give meaning attached to the statements given by the respondent, meaning
attached to other people or environment. (Example: Research on Hindu Muslim relation in
Muzaffarnagar district post riots.)
In the Qualitative research method, the researcher spoke to few Muslims in three different
localities, mixed locality, Muslim minority locality and also in Muslim majority locality.
The statements given by the respondents in these three different situations may be somewhat
different and will have some impact of the situation and the people living in the locality.
8
Analysis of Data in Qualitative Research
For recorded interviews, it has no meaning if the researcher reproduces the entire interview or
speech in his/her research.
The emphasis should not only be on what respondents said, but also on the pause, hesitation, where
respondent slows down. The researcher has to look for meaning of those pauses, hesitation, slowing
down, keeping mum at times etc.
The focus is not on how many respondents were interviewed but on the quality of the interview, on
the variety of meanings, attitudes, perceptions, interpretations attached to each statement.
Limitations:
(1) Very often the analysis of Qualitative data is subjective.
(2) The same data could be interpreted considerably different by different researchers
9
Analysis of Quantitative Data: How and why
The Quantitative Research encounters the issue of Data analysis much more compared to the
Qualitative method. Here technology comes into play.
The data collected in Qualitative research can be analyzed without the use of computer, but the data
collected in Quantitative study possibly cannot be analyzed without the use of computer.
Why is computer (some data analysis software) essential for Quantitative Data Analysis?
Surveys is always about big numbers, so one has to handle large number of responses. Such large
number of responses cannot be analyzed without use of computer. Necessary for getting fast results.
Useful for finding correlation between two or more variables
It is also useful for more sophisticated analysis, Correlation, Regression, T-Test, ANOVA etc.
It is also useful for data storage
10
Steps in Analysis of Quantitative Data
All responses obtained through interviews whether using open ended or closed ended questions must
be coded.
What is Coding?
It is the process of converting the written, descriptive responses into numerical.
The next step in data analysis is checking the coded data.
After coded data is checked, it is entered into the computer using different software.
Different packages for data analysis
Excel, SAS,SUDAAN, SPSS
How does these packages work?
The very first analysis of any statistical data is the frequencies.
Next step is to look at correlations between two or more variables, by running cross table, regression
or correlation .Useful in response classification , categorization Calculating the missing responses.
11
Correlation
Correlation: The linear relation between two are more variable is known as correlation. Correlation analysis is
used to measure strength of the association (linear relationship) between two connected variables. It can also
disclose the directions of sympathetic movements in variables.
Positive or direct correlation
If the two variables increase or decrease together in the same direction, they have a positive correlation.
e.g. The height and weight of group of person. The yield of crap and amount of rain fall.
Negative or inverse correlation
If, increases in one variable are associated with decreases in the other, they have a negative correlation.
e.g. The price and demand of a commodity. The volume and pressure of prefect gas.
Properties: (i) it is pure number.
(ii) It is independent of unit.
(iii) It is independent of change of origin and scale of data.
(iv) It lies between -1 to +1.
(v) It is zero order.
12
Regression
Regression: The functional relation between two are more variables is known as regression.
Regression analysis is used to:
(i) Predict the value of a dependent variable based on the value of independent variable
(ii) Explain the impact of changes in an independent variable on the dependent variable
13
Population Linear Regression
y β0 β1x ε
Variable
14
Formulation and Testing of Hypothesis
What is Hypothesis ?
Meaning and definition
•Any statement about population or population parameters is known as hypothesis. It may be true or
may not be true.
•A research hypothesis is a formal statement about the population or expectation about the outcome
of a study.
•A tentative and a testable statement about a relationship involving two or more variables is known
as hypothesis.
Confusion/Question/ Problem/Research/Thesis
•What is a research question.
•How does it link to the literature review and conceptual framework.
•Link between a research question and a hypothesis.
•Types of research questions.
•ladder of research questions.
•Solutions makes thesis.
15
Formulation and Testing of Hypothesis
Nature of Hypothesis
• A hypothesis “ It may prove to be correct or incorrect ”.
• In any event, however, it leads to an empirical test .
• A hypothesis is not held to be true until proven.
• When proven a HYPOTHESIS becomes a THESIS.
Concept of Hypothesis
Theoretical construct
Problem
Hypothesis
Theoretical construct
Solution
16
Formulation and Testing of Hypothesis
Need of hypothesis
• To Identify the data to be collected.
• To organize the data most effectively.
• To draw an inference about the population.
For example
(i) A given medicine cures 95% of the patient taking it.
(ii) A particular species of fish has life of 5 years.
(iii) In case of child birth, there is an equal chance of meal and female birth.
Sources of Hypothesis
• History.
• Analysis.
• Findings of other studies.
• Deduction from theories.
• Value-orientation of culture.
For example
(i) Intuition, hunches, knowledge,
(ii) curiosity, experience and other sources.
17
Formulation and Testing of Hypothesis
Characteristics of Hypothesis
• Empirically testable
• Closest to observable things
• Conceptually clear
• Specific
• Should have theoretical orientation
• Power of prediction
• Related to available measuring techniques
19
Formulation and Testing of Hypothesis
Null Hypothesis (H0): No significant difference
Difference between observed and expected is zero. H0, is the claim that is initially assumed to be true.
According to Professor R.A. Fishier: Null hypothesis is the hypothesis which is tested for possible rejection,
Under the assumption that is true.
Alternative Hypothesis (H1) Difference between observed and expected is not zero.
H1, is the assertion that is contrary to H0 .Any hypothesis which is complementary to the null hypothesis is
known as alternative hypothesis.
Test are basically two type
Small sample test: Sample size (n) < 30.
Test of Significance, T-test and F-test
Large sample Test: Sample size (n) ≥ 30.
Test of Significance, Chi Square -test and Z-test
Types of errors in hypothesis testing
Type I error: Reject the null hypothesis when it is true,
Type II error: Accept the null hypothesis when it is wrong.
20
Formulation and Testing of Hypothesis
Procedure for testing of hypothesis:
(i) Null hypothesis: Set up the Ho.
(ii) Alternative Hypothesis: Set up the H1.
(iii) Level of significance: Choose the appropriate level of significance(1%, 5%, 10%) on the basis of risk in
sample.
(iv) Test Statistics: Compute test hypothesis.
(v) Conclusion: Compare the computed value and calculated value.
• If computed value > Tabulated value we reject Ho.
• If computed value < Tabulated value we accept Ho.
21
Hypothesis testing
T-test [William Gosset (1908)]
(x )
t
s/ n
• T-test is small sample test (<30)(not<5).
• It is called student-t test (pen name).
• Degree of freedom (n-1).
• All observations of the sample is independent.
• Data are normally distributed.
• Population standard deviation is not known.
• Sample size is given (sample standard deviation is known)
• Range -∞ to ∞.
22
Hypothesis testing
Example based on T-Test
Problem: A manufacture of dry cell claimed that life of their cell is 24 hours. A sample of 10 cells had mean
life is 22.5. hours with a standard deviation of 3.0 hours. On the basis of available information, test whether the
claim of manufacturer is correct. [Given to.05,9=2.26]
Ttab. >Tcal
We accept the null hypothesis.
i.e. there is no significant differences.
23
Hypothesis testing
Z –test [R A Fisher(1923)]
(x )
z
/ n
• All observation of sample is independent
• Normally distributed
• Sample size sufficiently large(>=30)
• Population standard deviation is known
• Correlation coefficient is not zero
• Test for significance of single mean , difference between two means & its proportions.
• Z-test has only one critical points i.e. 1.96 for 5% in two tail test
24
Hypothesis testing
Solution: Assuming that the lifetime of cell is normally distributed. Here we test Ho: µ=10.2, Vs Ha: µ≠10.2,
n=400, S.D.=2.25.
( x ) (10 10.2)
Z 1.770
/ n 2.25 / 400
Ttab. >Tcal.
We accept the null hypothesis.
i.e. there is no significant differences.
25
Hypothesis testing
Chi-square Test [Karl Pearson (1920)]
(oi ei ) 2
2
ei
•Chi-square test is non-parametric test
•It can be used for test of goodness of fit
•It is sampling analysis for testing for significance of population variance
•It is used in SRS method lies between 0 to 1
•Sample size is large (n>50)
•Sample is independent, cell frequency linear
•Expected frequency will not same(e<5)
26
Hypothesis testing
Problem Based on Chi-square test.
A large electronic firm which hires many handicapped workers wants to determine whether their handicapped
effect such workers performance on the basis of sample data given below. Is it reasonable to believe that the
handicapped has no effect on workers performance?
Workers Above Average Average Below Average
Handicapped 37 113 32
No handicapped 23 97 28
How to perform chi-square test of in defence on the basis of a contigency table?
Solution:
O E (O-E) (O-E)^2 [(O-E)^2]/E
37 33 4 16 0.484
113 116 -3 9 0.077
32 33 -1 1 0.030
Chi- Square Cal= 0.591, Chi- Square Tab=5.991
Chi- Square Tabulated > Chi- Square Calculated
Null hypothesis is accepted.
27
Hypothesis testing
F-test
n1 2
s1
S12
n1 1
F 2
S2 n2 2
s2
n2 1
It is ratio of two chi- square variates with their respective degree of freedom.
It gives comparison of two chi-square variate.
28
Hypothesis testing
Problem based on F-test
Problem: Pumpkin were grown the experimented condition two random samples off 11 and 9 pumpkin. Show
that sample standard deviation of their weight as 0.8 and 0.5 respectively assuming that the weight of
distribution test the hypothesis is that two variance are equal . That two agents the alternative that they are not
10% level.
F tab=3.5
F cal=2.5
We accept the null hypothesis.
29
How to Write
30
What is a Report?
A report is a very formal document that is written for a variety of purposes in all the disciplines.
• The final and the last step in any research is to make report of the findings to make the research
and the study accessible to all.
• By this report, the research sponsors, government, students, academicians can get to learn and
understand the research related work.
• Research report is considered a major component of the research study for the research task
remains incomplete till the report has been presented and/ or written.
• The purpose of research is not well served unless the findings are made known to others.
• The task of report writing must be accomplished by the researcher with utmost care and s/he may
seek the assistance and guidance of the experts.
31
Language of Research Report
Minimum Requirements
• Clarity : Confusing and ambiguous statements to be avoided
• Precision: Generalities and vague statements to be avoided
• Appropriacy: Comprehensive not only to academics but to all intended readers
• Completeness: All issues should be given adequate emphasis
• Objectivity: Subjectivity and emotionality should be avoided
• Fairness: Presentation of findings in a fair and unbiased manner
• Verifiability: Information presented must be readily verifiable
• Impersonality: Report should be presented in the third person
32
Language of Research Report
Writing Style
• Poorly written report, difficult to understand is of no value
• Neatly typed, free of spelling and typographical errors
Clarity
• It is essential to be precise and clear
• Think about intended audience
• Presentation in orderly and logical sequence
• For producing a more organized report-use an outline
• After completing first draft- carefully read (grammar & spell checks)
• Get feedback from others on report- ask them to read critically and suggest improvements
• Write several drafts till you have satisfactory final product
33
Writing
34
Structure of Report
Title of Report :Concise heading indicating what the report is about
Contents: List of major sections and headings with page numbers
Abstract/Synopsis : Concise summary of main findings
Introduction :Why and what you researched
Literature Review: (sometimes included in the Introduction) - Other relevant research in this area
Methodology : What you did and how you did it
Results: What you found
Discussion: Relevance of your results, how it fits with other research in the area
Conclusion : Summary of results/findings
Recommendations (sometimes included in the Conclusion) :What needs to be done as a result of your
findings
References or Bibliography :All references used in your report or referred to for background information
Appendices : Any additional material which will add to your report
35
References v/s Bibliography
• Bibliography is a list of all the Books, Journals, Monographs, Theses, Documents, Data Sets etc that we
consulted during our research work.
•It is not necessary that they be cited in the text of the report.
•References indicate that we have borrowed ideas, information, text etc. from those documents.
Bean, Susan (1981): “Towards a Semiotics of ‘Purity' and ‘Pollution’ in India”, American Ethnologist, 8(3), pp
575-95.
Beteille, Andre (1992): The Backward Classes in Modern India (Delhi: Oxford University Press).
Datta, Srilekha (2008): “Gunas (Qualities) in Nyāya-Vaiśeṣika Ontology” in PK Sen (ed.), Philosophical
Concepts Relevant to Sciences in Indian Tradition (New Delhi: CSC).
Pandian, MSS (2008): Brahmin and Non-brahmin: Genealogies of the Tamil Political Present (New Delhi:
Permanent Black).
Desai, IP (1976): Untouchability in Rural Gujarat (Mumbai: Popular Prakashan).
Dumont, Louis (1988): Homo Hierarchicus (New Delhi: Oxford University Press).
Foucault, Michel (1994): Archaeology of Knowledge (London: Routledge).
Guru, Gopal (1996): Dalit Cultural Movement in Maharashtra (Mumbai: Vikas Adhyayan Kendra).
36
Assessment Pattern
37
Applications
Ability to apply to scientific research process for the given research
problem including identify various sources of information for literature
review and data collection. Appreciate the components of scholarly
writing and evaluate its quality.
38
References
Bryman, Alan & Bell, Emma (2018): Business Research Methods (Fifth Edition),
Oxford University Press.
Rubin, Allen & Babbie (2016): Essential Research Methods for Social Work,
(Fourth Edition), Cengage Learning Inc., USA.
Kothari, C.R. & Gaurav G. (2019): Research Methodology (Fourth Edition), New
Age International Publishers.
Naresh Malhotra (2012): Marketing Research, (Fifth Edition) McGraw Hills.
39
THANK YOU
For queries
Email: vipin.e9155@cumail.in