Scribd
Upload
32 views13 pages

Ensuring Data Quality in Research

The document outlines the process of ensuring data quality through various stages including planning, collection, management, and analysis. It highlights the importance of error checking and prevention methods, such as good questionnaire design and training for staff involved in data handling. The ultimate goal is to produce clean, error-free data ready for analysis and presentation of results.

Uploaded by

kokozaw
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
32 views13 pages

Ensuring Data Quality in Research

The document outlines the process of ensuring data quality through various stages including planning, collection, management, and analysis. It highlights the importance of error checking and prevention methods, such as good questionnaire design and training for staff involved in data handling. The ultimate goal is to produce clean, error-free data ready for analysis and presentation of results.

Uploaded by

kokozaw
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd

Data Quality

Data needs (Variables)


Planning (Proposal)
Forms
Data collection

Raw data

Data management
Clean (error-free) data
Data analysis
Information
Data presentation
Results
Raw data

Data management

Clean (error-free) data

Data analysis (+ stat test)

Information

Data presentation

Results
Quality of data

 Planning the data needs of a study

 Data collection

 Data management (data processing)

 Data Analysis
Data Management

 (raw data in forms  clean data)

- coding, data entry, checking

- recoding missing values, creating new

variables

- file merging/splitting, file exporting


Data Management

 (raw data in forms  clean data)

- coding, data entry, checking

- recoding missing values, creating new

variables

- file merging/splitting, file exporting


Main Types of Errors in Data
Activities
Filling error handwriting, missing value

Copying error 17 56 0o

Transposition 3993

Coding error 12 02

Routing error Answers in the wrong order

Range error 3 in (1,2) 35 yr in(15-24yr)

Consistency Male , Parity - 1


error
Checking of Data Quality
 Manual checking (during data collection)
– by interviewer on completion of each interview
– by supervisor on each day of data collection
– by researcher periodically on samples

 Interactive checking (during data entry)

 Double entry and validation (after data entry)

 Batch checking (after data entry)


Correction of Errors

Raw Data Checking Clean data

Correction Errors

Omitting
Prevention of Errors

 Good questionnaire design


 Pre-coding
 Consistent coding system
 Training of interviewers & supervisors
 Training of data entry staff
 Collection and transfer of data collection forms
 Checking of error at all phases of study
– Checking during data collection
– Checking at data entry
– Checking after data entry
Plan for Data Management
 Number and type of data collection forms Quality

 Pre-coding/coding after data collection

 Training of interviewers, supervisors and data entry staff

 Manual/computer data entry (Epidta, EpiInfo, Access, Excel)

 Collection and transfer of data collection forms

 Checking for Data Quality


- Checking of errors during data collection
- Checking of errors at data entry
(check computer program & double entry)
- Checking of errors after data entry
(Cross-tab for all categorical variables &
summary statistics for all numerical variables
Summary

 To produce Clean Data ready for Analysis

 Checking of Errors at all steps of study

You might also like