Research Methodology
- 1. Research Methodology Unit III Descriptive Statistics Data preparation: editing, coding, classification and tabulation of data, Measures of central tendency, Probability concepts, Theoretical distributions: Binomial distributions, Normal distribution, and Poisson distributionData Analysis
- 2. Introduction to Data Processing Data processing is primary stage of data analysis. The quality of results obtained from the statistical techniques and their subsequent interpretation depends on how well the data were prepared & converted into a form suitable for analysis. Advantages of Data Processing- 1. Converts in a Suitable Format- In this process, efficiency of data will increase by transforming it into a suitable format i.e. computer readable. It will decrease the time taken in analysis. 2. Provides Accurate Data- In this process, researcher checks the accuracy of data. After proper data processing, all chances of errors in collected data are eliminated. 3. Provides Comprehensive Information- After completing this process, the raw data transforms into meaningful information that can be better understood & analysed efficiently by researcher. 4. Helps in Decision Making- The processed data helps the managers to make important decisions & take necessary actions.
- 3. Steps in Data Processing 1 • Validation 2 • Editing 3 • Coding 4 • Classification 5 • Tabulation
- 4. 1. Validation This step transforms raw data into validated data. In this step a researcher determines whether a survey interviews or observations were conducted correctly & free from bias. The validated data are then processed to produce final report. Phases of Data Validation 1. Field Validation- This is a check in the actual field of data collection. There are three sub points- a) Data Collection Process, b) Proper Screening & c) Detection of Fraud. 2. Validation within the Firm- This is the checking of data completeness & data usability within the research firm. There are two sub-steps; a) Data Completeness & b) Data Usability Important Guidelines for this Process- i) Check-Backs ii) Review the Questionnaire & the Interviewing Instructions iii) Evaluate the Reputation of the Interviewers
- 5. 2. Editing The second important step in data processing is editing. Since raw data has many errors in the collection. In this stage these errors are corrected. Again there are three sub-points; a) Completeness b) Accuracy c) Consistency Stages of Editing- a) Field Editing- editing during data collection b) Office/Central Editing
- 6. 3. Coding This is the third step of data processing. In this step, the researcher converts the data into meaningful categories & then assigning symbols to each of these categories. These symbols are nothing but codes. E.g. code ‘1’ is taken to married & ‘2’ for unmarried persons. Coding includes three basic activities, viz. formulating the categories, allocating answers to those categories & finally assigning codes to those categories. Principles of Coding 1) Relevance: The categories should be able to contain the data necessary to test hypotheses. 2) Exhaustiveness: It means all raw data should belong to some particular categories. Researcher often include category like, “Others”. 3) Mutual Exclusiveness: The meaning of this point is ‘no interdependence’. E.g. if question is, “Which dishes do you want to take- Dal Firai, Rice Rajama, fish-rice, chicken kadhai, etc., Many respondents will choose more than two options. Therefore, coding the data will not be mutually exclusive. This problem can be solved by coding viz. ‘1’ for veg.., & ‘2’ for non-veg.., 4) Uni- Dimensional: A category should be defined using a single dimension only. For example, If a data set defines the occupations like manager, engineer, teacher, doctor, etc. then for locating unemployed doctor is tough task. For this solution, we create as unemployed manager, employed manager, unemployed engineer, employed engineer etc.
- 7. Coding (cont…) Procedure of Coding- there are following stages for coding; 1. Identifying Open Coding: In this stage, the researcher tries to identify the possible open categories to locate data. For example; formulating the open categories is done by constantly comparing the data to incidents (different dishes) and incidents to categories (Pure Veg, Pure Non-Veg. Mixed..) 2. Axial Coding: This is next to open category, where a coding paradigm (systematic arrangement) is developed. For which researcher uses blocks & arrows for information. 3. Selective Coding: This is the final stage of coding process. In this stage the paradigm is refined & presented as theory of the complete process. This theory includes propositions that indicate the possible concept to be explored & tested in future.
- 8. 4. Classification The meaning of classification is to create homogeneous classes into which the edited & coded data can be grouped on the basis of their common characteristics. For example; classify 50 respondents into smokers (35) & non- smokers (15). Principles in Classification of Data- For effective classification, the following principles should be followed; 1) Unambiguous Classification- The category should be defined clearly. For example, in the classification of smoking habit, if there are only two classes viz. ‘smoker’ & ‘non-smoker’. This classification can confuse the researcher to locate the persons of ‘occasional smoker’. 2) Single Classification Principle- 3) Mutually Exclusive Categories- 4) Mutually Exhaustive Categories- 5) Relevant for Research Project- Types of Classification 1) Classification According to Attributes 2) Classification According to Class-Interval
- 9. 5. Tabulation This is the final stage of data processing. In this stage data are represented in the form of tables. Types of Tables A) Based on Number of Characteristics Used- On the basis of number of characteristics, we divide tables into two categories; a) Simple Table or One-Way Table- when there is only one characteristic, i.e. Class (MBA IIA, B & C) & one characteristic viz. No. of Students. b) Complex Tables- When table is prepared for more than one characteristics, complex tables are further divided into following sub-categories i) Two Way Table- when two characteristics are there. i.e. Class (MBA & PGDM) & Two Characteristics No. of Students in the first & second year. ii) Three Way Table- For three interrelated characteristics i.e. three characteristics as No. of students in the first & second year belonging to North & South region. iii) Multiple-Way Table- For example table for population census. Based on the Purpose of Study a) General Purpose Table- size will be large b) Special Purpose Table- size will be small
- 10. Tabulation (Cont…) Rules of Tabulation, A) Rules Regarding the Table Structure: 1) Table Number 2) Title 3) Number of Rows & Columns 4) Size of Columns Arrangement of Items 5) Units & Derivatives: like weight in Kg., Distance in Km., Population in crores, etc. 6) Source 7) Totals B) General Rules 1) Attractive Shape 2) Simplicity 3) Free from Irrelevant Data 4) Non-availability of Data: Write ‘x’ or ‘-’ for data not available Significance of Tabulation
- 11. Tabulation (Cont…) Significance of Tabulation: Some important advantages of tabulation are as follows; 1. Helps in Comparing Data 2. Important for Data Analysis 3. Gives a Bird’s Eye View 4. Simplification of Complicated Data 5. Maximum Presentation of Data 6. Gives Overview