Scribd Logo
Upload

Welcome to Scribd!

  • 273 views

    Data Reduction & Estimation

    Uploaded by

    Sumit Kumar Sharma
    This document discusses data reduction and estimation techniques used in research. It describes the data reduction process which involves 6 steps: field controls, editing, coding, transcribing, generating new variables, and calculating statistics. Field controls monitor data collection to minimize errors. Editing ensures data is present, readable, and accurate. Coding assigns data to established categories. Estimation techniques infer values about a population based on a sample.

    Copyright:

    Attribution Non-Commercial (BY-NC)
    Download as PPT, PDF, TXT or read online from Scribd

    Data Reduction & Estimation

    Uploaded by

    Sumit Kumar Sharma
    273 views12 pages
    This document discusses data reduction and estimation techniques used in research. It describes the data reduction process which involves 6 steps: field controls, editing, coding, transcribing, generating new variables, and calculating statistics. Field controls monitor data collection to minimize errors. Editing ensures data is present, readable, and accurate. Coding assigns data to established categories. Estimation techniques infer values about a population based on a sample.

    Original Description:

    research methodology

    Copyright

    © Attribution Non-Commercial (BY-NC)

    Available Formats

    PPT, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document discusses data reduction and estimation techniques used in research. It describes the data reduction process which involves 6 steps: field controls, editing, coding, transcribing, generating new variables, and calculating statistics. Field controls monitor data collection to minimize errors. Editing ensures data is present, readable, and accurate. Coding assigns data to established categories. Estimation techniques infer values about a population based on a sample.

    Copyright:

    Attribution Non-Commercial (BY-NC)
    Download as PPT, PDF, TXT or read online from Scribd
    Download as ppt, pdf, or txt
    273 views12 pages

    Data Reduction & Estimation

    Uploaded by

    Sumit Kumar Sharma
    This document discusses data reduction and estimation techniques used in research. It describes the data reduction process which involves 6 steps: field controls, editing, coding, transcribing, generating new variables, and calculating statistics. Field controls monitor data collection to minimize errors. Editing ensures data is present, readable, and accurate. Coding assigns data to established categories. Estimation techniques infer values about a population based on a sample.

    Copyright:

    Attribution Non-Commercial (BY-NC)
    Download as PPT, PDF, TXT or read online from Scribd
    Download as ppt, pdf, or txt
    of 12

    DATA REDUCTION & ESTIMATION

    Dr. CK. Dash

    Data reduction
    Data reduction refers to process of getting the data ready for analysis and the calculation of summarizing or descriptive statistics.

    Estimation techniques
    Estimation techniques involve inferring the value of some group (called a Population) from a subset of that group (a sample). Since virtually all research studies involve samples, estimation techniques are very important.

    Data reduction Process


    1. Field controls, 2. Editing, 3. Coding, 4. Transcribing 5. Generating new variables, 6. Calculating summarizing statistics The first five of these steps are concerned with developing a basic data array that is as complete and as error-free as possible. The last step involves calculation made from the array.

    1.Field control
    Field controls are procedures designed to minimize errors during the actual collection of data. These controls involves ensuring the sampling data collection and measurement tasks are carried out as specified. Some field controls require both monitoring & validation procedure. Re-contacting a sample of respondent (10-20%) from each interviewers list for conformation, Effective field controls and editing require several variables in addition to the variables required by the research problem.

    Editing
    The responsibility of the editor is to ensure that the data requested are present, readable and accurate. Many questionnaires are pre-coded and are entered directly into computer processing with little or no editorial analysis (save time & Money)

    Missing Data
    It is very common for a questionnaire to be returned with one or more specific question unanswered. (Known as item non-response) Unanswered questions are assigned a missing data code ( Blank, -9) Exclude from the sample or average value of the variable. Editor can have respondents recontacted to collect key bits of missing information.

    Ambiguous Answers
    Many questionnaires contain one or more response whose meaning is not clear. Exp: I almost never drink carbonated drinks, but I do usually have several.
    10-15 cokes and several fruit drinks, 12 if you cont mixed drinks, 6- Summer, 0- Winter Other kinds of ambiguities with which editors must deal are illegible responses and marks between response categories.

    Accuracy / Quality
    As editors review a series of questionnaires, they should not suspect responses. Respondents will sometimes rush through questionnaires in an almost random manner. This trends to produce a number of inconsistent responses such as a high-income, a low paying job categories, or un awareness of brand, i.e also reported as frequently used. It appears that the respondents were haphazard in completing them.

    Editing procedures
    Interview should be turned in and edited promptly, Editors should be assigned interviews by interviews by interviewer, Editors should make changes by crossing out or transferring data rather than by erasing When more than one Editor is used, editing instructions should be prepared. Editing requires extensive use of judgement.

    Coding
    Coding may also be done by the editor, it is a separate step, which involves establishing categories and assigning data to them

    Establishing Categories
    Categories for answers to multiple-choice or dichotomous questions are established at the time the question is formulated. Open-ended questions may also have response categories established at the time they are formulated.

    You might also like