Microsoft Word research method fm doc

Data processing – quantitative data 
Decide whether to process and analyse the data from questionnaires: 
• 
manually
, using data master sheets or manual compilation of the questionnaires, or
• 
by computer
, for example, using a micro-computer and existing software or self-
written programmes for data analysis.
Data processing in both cases involves: 
• 
categorising the data,
• 
coding, and
• 
summarising the data in data master sheets, manual compilation without master 
sheets, or data entry and verification by computer.
1. Categorising 
Decisions have to be made concerning how to categorise responses. 
For categorical variables that are investigated through closed questions or observation, the 
categories have been decided upon beforehand. 
In interviews the answers to open-ended questions (for example, ‘Why do you visit the health 
centre?’) can be pre-categorised to a certain extent, depending on the knowledge of possible 
answers that may be given. However, there should always be a category called ‘Others, 
specify . . .’, which can only be categorised afterwards. 
These responses should be listed and placed in categories that are a logical continuation of 
the categories you already have. Answers that are difficult or impossible to categorise may be 
put in a separate residual category called ‘others’, but this category should not contain more 
than 5% of the answers obtained. 
For numerical variables, the data are often better collected without any pre-categorisation. If 
you do not exactly know the range and the dispersion of the different values of these 
variables when you collect your sample (e.g., home-clinic distance for out-patients, or 
income), decisions concerning how to categorise and code the data at the time you develop 
your tools may be premature. If you notice during data analysis that your categories had been 
wrongly chosen you cannot reclassify the data anymore. 

---

Research methodology 
70
2. Coding 
If the data will be entered in a computer for subsequent processing and analysis, it is 
essential to develop a CODING SYSTEM. 
For computer analysis, each category of a variable can be coded with a letter, group of letters 
or word, or be given a number. For example, the answer ‘yes’ may be coded as ‘Y’ or 1; ‘no’ 
as ‘N’ or 2 and ‘no response’ or ‘unknown’ as ‘U’ or 9. 
The codes should be entered on the questionnaires (or checklists) themselves. When 
finalising your questionnaire, for each question you should insert a box for the code in the 
right margin of the page. These boxes should not be used by the interviewer. They are only 
filled in afterwards during data processing. Take care that you have as many boxes as the 
number of digits in each code.