•
SAMPLING ERROR VS NON SAMPLING ERROR
RESEARCH METHODOLOGY
RESEARCH METHODOLOGY
1.
SAMPLING ERRORS: IS ONE WHICH OCCURS DUE TO UNREPRESENTATIVE OF THE
SAMPLE SELECTED FOR OBSERVATION
2.
NON SAMPLING ERRORS: IS AN ERROR ARISE FROM HUMAN ERROR SUCH AS ERROR IN
PROBLEM IDENTIFICATION,METHODS OR PROCEDURES USED ETC
•
SAMPLING ERRORS
1.
DENOTES A STATISTICAL ERROR ARISING OUT OF CERTAIN SAMPLE SELECTED BEING
UNREPRESENTATIVE OF THE POPULATION OF INTEREST.
2.
IT IS AN ERROR WHICH OCCURS WHEN THE SAMPLE SELECTED DOES NOT CONTAIN THE
TRUE CHARACTERISTICS,QUALITIES OR FIGURES OF THE WHOLE POPULATION
3.
SAMPLING ERROR MEANS DEVIATION BETWEEN THE TRUE MEAN VALUE OF THE ORIGINAL
SAMPLE AND THE POPULATION
•
MAIN REASON:-
1.
SAMPLER DRAWS VARIOUS SAMPLING UNITS FROM THE SAME POPULATION BUT THE UNITS
MAY HAVE INDIVIDUAL VARIANCES
2.
THAT COULD BE DUE TO DEFECTIVE SAMPLE DESIGN,FAULTY DEMARCATION OF
UNITS,WRONG CHOICE OF STATISTIC,SUBSTITUTION OF SAMPLING UNIT BY THE ENUMERATOR
BY CONVENIENCE
•
IMPORTANT POINT
1.
SAMPLING ERROR IS A STATISTICAL ERROR THAT OCCURS WHEN AN ANALYST DOES NOT
SELECT A SAMPLE THAT REPRESENT THE ENTIRE POPULATION OF THE DATA
2.
RESULT OF THE SAMPLE DO NOT REPRESENT THE RESULTS OBTAINED FROM THE ENTIRE
POPULATION
3.
CAN BE REDUCED BY RANDOMIZING SAMPLE SELECTION AND OR INCREASING THE NUMBER
OF OBSERVATION
4.
SAMPLE ERROR CAN BE ELIMINATED WHEN THE SAMPLE SIZE IS INCREASED
•
NON SAMPLING ERROR
1.
IS A TYPE OF ERROR OCCURRING IN STATISTICAL INVESTIGATION IRRESPECTIVE OF
WHETHER THE INVESTIGATION IS BASED ON SAMPLE OR THERE IS COMPLETE ENUMERATION
2.
MAGNITUDE OF NON SAMPLING ERROR MAY BE MORE IN COMPLETE
ENUMERATION,ENUMERATING MUCH LARGER NUMBER OF UNITS
•
CAUSES OF NON SAMPLING ERRORS
1.
ERRORS IN COVERAGE BY THE USE OF DEFECTIVE FRAME/WRONG DEFINITION/INADEQUATE
DATA SPECIFICATION
2. ERRORS DUE TO OBSERVATION: RESPONSE AND MEASUREMENT ERRORS
3.
ERROR DUE TO NON RESPONSE OR NON OBSERVING THE DATA
4.
ERRORS IN PROCESSING AND ANALYSIS OF DATA
5.
ERRORS IN PREPARATION OF REPORT OR DISSEMINATION OF FINDING
6.
CAN NOT BE EFFECTIVELY ESTIMATED BUT MAY BE CONTROLLED TO REASONABLE EXTENT
•
OBSERVATION ERRORS
•
IN FIELD SURVEY,THE OBSERVATION ERRORS OCCURS WHEN OBSERVATION MADE ARE NOT
TRUE VALUE OF THE OBSERVATION.
•
ERRORS OF OBSERVATION :-
•
MEASUREMENT ERRORS:-FAULTY MEASURING DEVICES ,INEFFICIENCY OF THE INVESTIGATORS. REASONS FOR SUCH ERRORS:-
1. INADEQUATE INSTRUCTIONS TO FIELD
2.
INADEQUATE SUPERVISION OF ENUMERATORS
3.
INADEQUACY TRAINED AND INEXPERIENCED FIELD STAFF
•
RESPONSE ERROR:-
1.
INACCURATE ANSWERS GIVEN BY THE
2.
MISINTERPRETATION OR RECORDING WRONGLY
•
CONSIST OF
1.
RESEARCH ERRORS
2.
RESPONDENT ERROR
3.
INTERVIEWER
§ NON RESPONSE ERROR
•
RESPONSE ERROR
•
RESEARCH ERRORS:-
1.
SAMPLING ERRORS
2. MEASUREMENT ERRORS
3.
DATA ANALYSIS ERRORS
4.
POPULATION DEFINITION ERRORS
•
RESPONDENT ERROR:-INABILITY ERRORS AND UNWILLINGNESS ERROR
•
INTERVIEWING ERROR:-
1.
QUESTIONING ERROR
2.
RECORDING ERROR
3.
RESPONDENT SELECTION
•
DIFFERENCE BETWEEN SAMPLING AND NON SAMPLING
ERROR
•
SAMPLING ERROR
- IS A STATISTICAL ERROR HAPPENS DUE TO THE SAMPLE SELECTED DOES NOT
PERFECTLY REPRESENT THE POPULATION OF THE INTEREST
- BECAUSE OF THE VARIATION BETWEEN THE TRUE MEAN VALUE FOR THE SAMPE AND
THE POPULATION
- SAMPLING ERRORS OCCURS IN THE RANDOM SAMPLE ONLY
- ARISES ONLY WHEN THE SAMPLE IS TAKEN AS A REPRESENTATIVE OF POPULATION
- MAINLY ASSOCIATED WITH THE SAMPLE SIZE. AS THE SAMPLE SIZE INCREASES
THE POSSIBILITY OF ERROR DECREASES
•
•
NON SAMPLING ERRORS
1.
NON SAMPLING ERRORS OCCURS DUE TO SOURCES OTHER THAN THE SAMPLING WHILE
CONDUCTING SURVEY ACTIVITIES
2.
NON SAMPLING ERROR ARISES BECAUSE OF DEFICIENCY AND INAPPROPRIATE ANALYSIS
OF DATA
3.
CAN BE RANDOM OR NON RANDOM
4.
ARISES BOTH IN SAMPLING AND COMPLETE ENUMERATION
5.
IS NOT RELATED TO SAMPLE SIZE SO WITH THE INCREASE IN SAMPLE SIZE IT WOULD
NOT BE REDUCED.
No comments:
Post a Comment