Research Process

Prof .P. G. Tadasad

Source: http://www.informedbusinessdecisions.com/RoadMapt400c.jpg

2
 Research Design
 Overall plan or programme of research
 General blueprint for the collection, measurement and analysis of data
 An outline of what an investigator will do from day one till he/she submits
the report
 Acts as a standard guidepost
 Helps in carrying out research validly, objectively, accurately, and
economically
 Forecasts the possibility of fruitless enquiry and acts as an insurance
against future failure

3
 Uses
 Provides answers to
 What is the object of study?

 What data are needed? And how will these be gathered?

 What will be the time, space and sample of study?

 How will data be analysed and interpreted?

 What will be the cost and labour required

 What will be the level of accuracy desired in the results?

4
 Types of research design
 Co-relational
 Explores or tests relations between variables
 Field (survey)
 Studies participants in their natural setting; Maximizes realism
 Experimental

Directly establishes cause-effect nature of relationship between variables
 Qualitative
 Non-quantitative , Not necessarily informal data collection
 Meta-analysis
 Statistically combines results of existing research to estimate overall size of relation
between variables

5
 Parts of Research design
 Sampling design
 Describes various sampling methods to be used for selecting subjects for

study
 Observational design
 Describes the way in which observations are to be made

 Statistical design
 Describes the statistical tests to be used for analysis and interpretation of

data
 Operational design
 Describes how the entire operation of research is to be carried out

6
 Sampling
 Much emphasis in Agricultural research
 Economy in expenditure
 Speed, Scope, accuracy, Practicability
 Sampling process
 Define population
 Sampling frame
 Specify sampling unit
 Specify sampling method
 Determine sampling size
 Specify sampling plan
 Select the sample
 Sampling methods
Random or Probability sampling Non-random or Non-probability samples
 Simple random  Purposive sample
 Systematic sample  Quota sample
 Stratified random  Convenience sample
 Multi-stage random
 Cluster sample or area sample
 Sequential sample
 Replicated sample
 Random Number Tables
 Tippett’s tables
 Fisher and Yates tables
 Kendall and Babington Smith tables
 Rand Corporation Tables
 Rao, Mitra and Mathai tables
 Determination of sample size
 Resources available
 Nature of the study
 Method of sampling followed
 Nature of population
 Degree of accuracy required
 Degree of confidence desired
 Sampling and Non-sampling errors
Sampling error:
 Difference between the statistic Non-sampling errors
and the parameter (Duckworth –  Defective frame and faulty
Lewis Rule) sampling

New Zealand losing in World Cup
 South Africa losing in World cup
 Inappropriate methods of data
 Pakistan air crash collection, coding, tabulation

Chandrayan –II  Inadequate supervision
 COVID-19 test results  Incomplete coverage
 Measurement and Scales
 Measurement  Scales
 Nominal  Rating scales
 Ordinal  Attitude scales
 Interval
 Ratio
 Nominal measurement
 Determination of equality
 Helps in grouping or classifying
 Percentage and Mode
 Chi-square and Binomial test
 Ordinal Measurement
 Determination of greater or lesser
 Rankings, Preference data, market position, attitude
measures, psychological measures
 Median
 Rank-order correlation
 Interval measurement
 Determination of equality of intervals
 Index numbers, Attitude measures
 Mean, range, standard deviation
 Product-moment correlation, t-test, Factor analysis
 Ratio measurement
 Determination of equality of ratios
 Sales, units produced, Number of customers, costs,
age
 Geometric mean
 Coefficient of variation
 Rating scales
Discrete scale:
 Two or more major categories are
Continuous or graphic scale:
provided representing discrete  On an interval scale which
amounts of some characteristics is continuous
 Extremely good, Above
 Highly satisfied
average, Average, Below  Highly dissatisfied
average, Poor
10 9 8 7 6 5 4 3 2 1
 Attitudinal scale
 Lickert’s summated scale
 Thurstone’s Equal Appearing Interval Scale
 Guttman’s Cumulative Scale
 Osgood’s Semantic Differential Scale
 Data Processing
 Decision regarding mode of data processing
 Manual
 Software packages
 Data validation
 Data classification
 Transcription
 Coding
 Data Entry
 Graphical representation
 Data presentation (Use of visual aids)
 Tables  Schedule charts
 Line Graphs  Flow charts
 Bar graphs  Decision charts
 Pie Charts  Illustrations
 Troubleshooting charts  Circuit diagrams
 Organizational charts  Photographs
 Gant Charts  Drawings
LET US CHECK YOUR UNDERSTANDING…..
1. The correct sequence of stages of research is…….
(A) Problem selection , data collection, statement of hypothesis
(B) Data collection ,Problem selection , statement of hypothesis, data
analysis
(C) Problem selection , hypothesis statement , data collection , data
analysis
(D) Hypothesis statement , Problem selection , data collection , data
analysis
2. In the following question, a statement is followed by an assumption. Consider the
statement and the following assumption and point out the correct Assumption.
 Statement (S): The preparation of research design involves no decision
making on techniques to be employed.
 Assumption (A): the purpose of research design can be broadly classified in
to four categories i.e., exploration, description, diagnosis and
experimentation.
 Find the correct answer from the codes given below:
(A) Both (S) and (A) are true and (A) is the correct explanation of (S).
(B) Both (S) and (A) are true and (A) is not the correct explanation of (S).
(C) (S) is true and (A) is false.
(D) (S) is false and (A) is true.
3. Which technique is generally followed when the
population is finite?
(A) Area Sampling Technique
(B) Purposive Sampling Technique
(C) Systematic Sampling Technique
(D) None of the above
4. Which one is called non-probability sampling?
(A) Cluster sampling
(B) Quota sampling
(C) Systematic sampling
(D) Stratified random sampling
5. Which of the following sampling method is
appropriate to study the prevalence of AIDS
amongst male and female in India in 1976, 1986,
1996 and 2006?
(A) Cluster sampling (B) Systematic sampling
(C) Quota sampling (D) Stratified random sampling
Thank you