Primary Data

12/4/2008 9:01:48 AM

COLLECTION OF DATA

Collection of Primary Data a) Observational Method b) Interview Method c) Questionnaire Method d) Schedules e) Other Methods

Interview method—Personal Interview Telephonic Interview

Schedule Types --- -a) Observation Schedule b) Rating Schedule c) Document Schedule d) Institution Schedule e) Interview Schedule Probability TYPES Simple Random Sampling Stratified Systematic Multi Stage Multi Phase Interpenetrating

(Procedure)

Define the Population Select the sampling Frame Specify the sampling Unit Specify the sampling Method Specify the sampling Size Specify the sampling Plan Select the Sample

RESEARCH DESIGN

Parts

Structured Observation Un structured Observation Controlled Observation Un Controlled Observation Participant Observation Non Participant Observation Disguised Observation

a) structured Interview b) Unstructured Interview

Questionnaire ---- Open Ended Questions Closed Ended Questions

SAMPLING

a) b) c) d) e) f) g)

a) Fill in the blanks b) Dichotomous c) Ranking Scale d) Multiple Choices e) Rating

Non Probability Quota Snow Ball Judgmental Accidental

a) Sampling b) Observational c) Statistical d) Operational

Types 1) For Exploratory Research

2) For Descriptive Research

i) Search in concerned Literature ii) Experience Survey iii) Analysis of ‘insight-stimulating examples’ i) Formulation of objective of study ii) Designing the data collection method iii) Processing & analyzing the data iv) Reporting the analysed data

1

Page 1 of 5 PDF created with pdfFactory Pro trial version www.pdffactory.com

12/4/2008 9:01:48 AM

ATTITUDE TESTING:

Attitude is the mental state of an individual to act or respond against or in favor in a particular situation or an object. Scaling:

Assigning numbers ( numerical ranking) to the situation indicative of an individual attitude according to a rule.

Data Analysis Scales: Nominal- It is least powerful scale, Used for categorizing, personal variables. Categorization of variable into several subclasses by assigning numbers, but these numbers can not be treated for arithmetic treatments. Ordinal- Various homogeneous groups of the data are formed between assigning ranks, these ranks might be based on one or more characteristics of population. Rank order rating, paired comparison can be used for it. Interval- It is useful for showing/measuring the interval with its natural variation like temperature. Ratio – For measuring properties for those qualities which have a natural zero Scale/point. It is the strongest scale. Scale Nominal Ordinal Interval Ratio

Order X √ √ √

Distance X X √ √

Origin X X X √

Scale decision / construction technique--a) Find out suitability of scale b) Find out necessary statistical analysis c) Find out suitable scale construction technique d) Selecting physical layout of scale e) Determining the scale categories that need to be used Multiple items scaling- a) Likert’s Summated rating scale b) Thrustone’s Equal appearing intervals scale c) Guttman’s Cumulative Scale Semantic Differential Scale – Factors Used

Type of multiple item scaling:

a) Evolution = Good/Bad b) Potency = Hard/Soft c) Activity= Slow/Fast

i) Pair wise comparison scale ii) Rank order scale iii) Constant sum scale iv) Rasch model scale v) Continuous rating scale vi) Likert scale 2

Page 2 of 5 PDF created with pdfFactory Pro trial version www.pdffactory.com

12/4/2008 9:01:49 AM

Definition of Research: “Research is a systematized effort to gain knowledge”- Redman & Mory Scope of Research:

Finance- Budgeting-Investment Purchasing- Procurement, production Management Distribution-Size of warehouse, out lets, Distribution policy Personnel Management- Recruitment, Job Designing Selection of suitable person on minimum salary etc.

Characteristic of Research:

a) b) c) d) e)

Types of Research:

Fundamental Applied Descriptive Historical Exploratory Experimental Ex-Post-facto Case study method

a) b) c) f) g) h) i) j)

Continuing Process Economic in Nature Methodical Process Interdisciplinary team approach Objective approach

a) Qualitative ResearchClassified as a) Experimental- for establishing cause- effect relationship B) Simulation- For building models, understanding future conditions c) Inferential- Understanding relationship D) Field- Going into field for observation b) Quantitative Research

E) Phenomenological- experience of a long period of time.

PROCESS OF RESEARCH: a) b) c) d) e) f) g) h) i) j) k) l) m) n)

Selecting the Topic Defining the research Problem Mention the research objectives Survey existing literature Development of working Hypothesis Preparing research design Exploratory Research/Conclusive Research Sampling design Data Collection Execution of Research Analysis of Data Hypothesis Testing Interpretation Preparation of Research Thesis. 3

Page 3 of 5 PDF created with pdfFactory Pro trial version www.pdffactory.com

12/4/2008 9:01:49 AM

Hypothesis: An approximate assumption that a researcher wants to test for its logical or empirical Consequences / Is an uncertain statement that involves the proposed answer to the problem. Types of Hypothesis: a) Null Hypothesis- A Statement of “No chance” or “No difference” or “no relationship”. Is formulated for possible rejection or nullification. b) Alternative Hypothesis- A logical opposite of” null-hypothesis”. c) Research Hypothesis- Test for which research is conducted. Approaches to testing of Hypothesis: 1) Classical- Based on analysis of sampling of data 2) Bayesian- Sampling plus data characteristics Test of Significance: a) Parametric

b) Non Parametric

Z- test t- test F- Test

Chi-square test Sign Test Median test Mann-Whitney U test Wilcox in-Matched pair test Kol-mogorov-Simrnov (KS) test

For Normal Distribution: a) One Tailed test- The rejection region is located on only one side of the normal curve, either left side or right side. b) Two Sided test - The rejection region is located on both the sides of the normal curve. Decision of which sided rejection area testIt is to be followed/used depends up on the research hypothesis (Ha). When Ha predicts high valuerejection is located on the right side of the tail and, when the Ha predicts lower value the rejection region is located n the left hand side of the curve. Steps in testing hypothesis: 1) 2) 3) 4) 5) 6) 7) 8) 9)

State the null hypothesis State the Alternative hypothesis State the appropriate statistical test ( from-F, t, Z) State the Level of Significance Compute the Statistics Selection of the Correct type of distribution-( Normal/Binomial/t-distribution/F-distribution) Compute the significance test level Obtain the critical value Make the decision 4

Page 4 of 5 PDF created with pdfFactory Pro trial version www.pdffactory.com

12/4/2008 9:01:49 AM

Types of Research Design: a) Design in exploratory Studies b) Design in descriptive Studies c) Design in Causal Studies Types (Originates) of Alternate Hypothesis: 1) Subject effect 2) History effect 3) Maturation effect 4) Demand effect 5) Testing effect 6) Regression to the mean test 7) Instrumentation 8) Halo effect 9) Mortality effect 10) Expectancy effect Experiment Design Types 1) Pre-Experimental (Pilot-test) ---- One shot Case study One Group Pretest Static group Comparison study 2) Quasi-Experimental-----------------Time Series Multiple Time Series Static group designs 3) True Experimental-------------------Post Test Equivalent Group Pre Test Post test Equivalent Group

Sampling Parameters: 1) 2) 3) 4) 5)

Universe: Total number of elements Sample: Representative units from the population Sampling Plan: The process/Technique followed for sample selection Statistics& Parameters: Sample mean is statistics, Population mean is Parameter of the Sample Sampling Error: Due to sample the minimum error that might be there, Larger the sample size lower will the error. 6) Significance level: The level of likelihood confidence about the happening of the events. 7) Precision: The tolerance within which the exact answer/out come will lie.+- range 8) Sampling distribution: F distribution, t distribution, Z distribution TYPES OF RESEARCH: A) 1) 2) 3) 4) 5)

Applied Research-------------- B) Pure Research Exploratory Research Descriptive Diagnostic Evaluation Action 5

Page 5 of 5 PDF created with pdfFactory Pro trial version www.pdffactory.com