Scribd
Upload
23 views

Structure Checking and Verification

This document discusses the process for structure checking and verification of MICS survey data. Structure checking involves checking that the number of questionnaires in the data file matches the cluster tracking form. Verification involves having a second data entry operator enter the same data and then comparing the two entered datasets to identify any differences. Differences identified are investigated by the data entry operators and corrections are made to both datasets until they are fully reconciled.

Uploaded by

Murugan Paramasivam
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
23 views

Structure Checking and Verification

This document discusses the process for structure checking and verification of MICS survey data. Structure checking involves checking that the number of questionnaires in the data file matches the cluster tracking form. Verification involves having a second data entry operator enter the same data and then comparing the two entered datasets to identify any differences. Differences identified are investigated by the data entry operators and corrections are made to both datasets until they are fully reconciled.

Uploaded by

Murugan Paramasivam
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 14

MICS Data Processing Workshop

Structure Checking and Verification

Structure Checking Flow Chart


Main Data Entry
Data Entry Operator 1

Structure Check
DP Supervisor

Correct Main Data File Data Entry Operator 1

Check ok? Yes

No

Investigate Errors DP Supervisor

Verification Data Entry


Data Entry Operator 2

Structure Checking

Check the number of questionnaires in the data file matches number of questionnaires on the cluster tracking form Check household information panel counts of women and under-fives (HH12-15) match number of questionnaires in the data file

The Cluster Tracking Form


Allows supervisor to organize and track data entry progress Questionnaire totals entered by questionnaire administrator Other information entered by supervisor

Cluster Tracking Form


Number of questionnaires Cluster number 101 Date received 23/08/05 Households Selected 17 Household 17/14 Womens 19/19 Childrens 21/19 V: Alpha M: Mariam 102 21/08/2005 17 17/16 23/19 18/16 V: Joy M: 103 21 V: M: 104 105 29/08/2005 23 V: 15 15/15 14/11 13/12 M: Alpha 7 29/08/05 3 23/08/05 7 4 25/08/05 21/08/05 24/08/05 25/08/05 25/08/05 27/08/05 27/08/05 Operator name M: Sarah Data entry Operator number 1 Creation date 23/08/05 26/08/05 28/08/05 28/08/05 Date check complete Date verification complete Date of raw data back-up Date of editing Date of final back-up

Cluster Tracking Checks


Check questionnaires in data file equal questionnaires on cluster tracking form Check done for:

Complete, Incomplete and Total Households Complete, Incomplete and Total Women Complete, Incomplete and Total Children

Cluster Checks

Womens questionnaires = HH12 Completed womens questionnaires = HH13 Under-five questionnaires = HH14 Completed under-five questionnaires = HH15

Sample Output - Cluster


MICS Data Structure Check Cluster: 3 Households | Women Total Comp Incomp | Eligible Interviewed | HH12 FOUND HH13 FOUND 2 1 1 | 5 5 4 4 | Children | Eligible Interviewed | HH14 FOUND HH15 FOUND | 4 4 3 3

Compare totals to those on cluster tracking form If totals different, check clusters questionnaires and resolve the difference

Household Checks

Womens questionnaires = HH12 Completed womens questionnaires = HH13 Under-five questionnaires = HH14 Completed under-five questionnaires = HH15

Sample Output - Household


MICS Data Structure Check Household: 1 Result: 1 Women | Children Eligible Interviewed | Eligible Interviewed HH12 FOUND HH13 FOUND | HH14 FOUND HH15 FOUND 4 4 3 3 | 2 2 1 1

Use listing of households to identify source of cluster level problems

Verification Flow Chart


Verification Data Entry
Data Entry Operator 2

Verification
DP Supervisor

Correct Both Data Files Data Entry Operator 1 & 2

Differences? No

Yes

Determine Correct Values Data Entry Operator 1 & 2

Backup Raw Data File


DP Supervisor

Verification

Purpose: Eliminating data entry errors Method:


1. program compares main and verification data files 2. program produces list of differences 3. data entry operators use listing and questionnaires to determine corrections 4. Corrections marked on listing 5. BOTH data files corrected by original operators 6. repeat 1-5 until no differences in listing!

A Differences Listing
Input File: C:\MICS\CSPRO\RAW\HH003M.DAT Reference File: A:\HH003V.DAT ----------------------------------------------------------------------------------------------Case Id Item Input File Reference File ----------------------------------------------------------------------------------------------[003001 ] ED3(6) ED3B(6) DA11(9) [00300404] IM1 IM2 IM2D IM2M IM2Y

202 02 2

204 04 1

1 31032004 31 03 2004

Verification Application

The file compare.cmp defines which items (i.e., variables) are to be excluded It is strongly recommended that this file not be changed

You might also like