Menu
Home
Log in / Register
 
Home arrow Environment arrow Research Methods in Anthropology: Qualitative and Quantitative Approaches
Source

Cleaning Data

No matter how careful you are, you’re going to make mistakes in coding and entering data into a computer. Even if you code behavioral data in the field directly into a handheld computer, believe me, you’ll make mistakes. Everyone does. The trick is to catch the mistakes.

The best way to catch coding and data entry errors is with a data editor, like Excel® or the spreadsheet in full-featured statistical analysis packages (see appendix E). Suppose you have a variable called NUMCHILD, or ‘‘number of children.’’ If you suspect that none of your informants have more than 10 children, you can tell the editor to ‘‘find the cases in the column for NUMCHILD where the entry is greater than 10.’’ If the program finds such cases, you can decide whether the entry is an error or if it just means the informant has a lot of kids. Similarly, if an informant says that she has never had children, you want to make sure that the variable for number of grandchildren is also 0.

You can move columns of numbers around with a data editor. When you enter data, you often don’t know which variables to put together. Later, as you become more familiar with your data, you’ll want to move columns around so you can eyeball the data matrix and look for mistakes and incompatibilities. Once you have your data into the computer and cleaned, you’re ready to start analyzing them (Further Reading: cleaning data).

 
Source
Found a mistake? Please highlight the word and press Shift + Enter  
< Prev   CONTENTS   Next >
 
Subjects
Accounting
Business & Finance
Communication
Computer Science
Economics
Education
Engineering
Environment
Geography
Health
History
Language & Literature
Law
Management
Marketing
Mathematics
Political science
Philosophy
Psychology
Religion
Sociology
Travel