Microsoft Word research method fm doc

What should the plan include? 
When making a plan for data processing and analysis the following issues should be 
considered: 
• 
Sorting data,
• 
Performing quality-control checks,
• 
Data processing, and
• 
Data analysis. 
 
Sorting data 
An appropriate system for sorting the data is important for facilitating subsequent processing 
and analysis. 
If you have different study populations (for example village health workers, village health 
committees and the general population), you obviously would number the questionnaires 
separately
. 
In a comparative study
it is best to sort the data right after collection into the two or three 
groups that you will be comparing during data analysis.
For example
, in a study concerning the reasons for low acceptance of family 
planning services, users and non-users would be basic categories; in a case-
control study obviously the cases are to be compared with the controls. 

---

Research methodology 
68
It is useful to number the questionnaires belonging to each of these categories separately 
right after they are sorted.
For example
, the questionnaires administered to users of family planning 
services could be numbered U1, U2, U3, etc., and those for the non-users N1, 
N2, N3, etc. 
Performing quality control checks 
Usually the data have already been checked in the field to ensure that all the information has 
been properly collected and recorded. Before and during data processing, however, the 
information should be checked again for completeness and internal consistency. 
If a questionnaire has not been filled in completely you will have MISSING DATA for some of 
your variables. If there are many missing data in a particular questionnaire, you may decide 
to exclude the whole questionnaire from further analysis. 
• 
If an inconsistency is clearly due to a mistake made by the researcher/research 
assistant (for example if a person in an earlier question is recorded as being a non-
smoker, whereas all other questions reveal that he is smoking), it may still be possible 
to check with the person who conducted the interview and to correct the answer.
• 
If the inconsistency is less clearly a mistake in recording, it may be possible (in a small 
scale study) to return to the respondent and ask for clarification.
• 
If it is not possible to correct information that is clearly inconsistent, you may consider 
excluding this particular part of the data from further processing and analysis as it will 
affect the validity of the study. If a certain question produces ambiguous or vague 
answers throughout, the whole question should be excluded from further analysis. 
(Normally, however, you would discover such a problem during the pre-test and 
change the wording of the question.)
For computer data analysis, quality control checks of data must also include a verification of 
how the data has been transformed into codes and subsequently entered into the computer. 
The same applies if data are entered into master sheets. 

---

Research methodology 
69