Microsoft Word research method fm doc

For example
, the straightforward answers of the smoking questionnaire for male smokers 
could be processed as follows: Master
 sheet for smokers (males)
Note that for age and number of cigarettes smoked both the raw data and the categories 
have been entered. This makes it easier to control for coding mistakes and allows for 
calculating averages. There are 31 male smokers; if there are less than 31 answers, there 
must be some non-responders (NR), as happened in Q9, or a mistake was made. If you work 

---

Research methodology 
72
with two persons, one reading and one writing, the risk of mistakes will be reduced, as you 
can discuss the answers and control for mistakes while filling in the data. 
(2) Compilation by hand (without using master sheets)
When the sample is small (say less than 30) and the collected data is limited, it might be 
more efficient to do the compilation manually. 
Certain procedures will help ensure accuracy and speed. 
1. If only one person is doing the compilation use manual sorting. If a team of 2 persons 
work together use either manual sorting or tally counting.
2. To do manual sorting the basic procedure is to:
o
Take one question at a time, for example, ‘use of health facility’,
o
Sort the questionnaires into different piles representing the various responses 
to the question, e.g., hospital/ health centre/ traditional practitioners) and
o
Count the number in each pile.
When you need to sort out subjects who have a certain combination of 
variables (e.g. females who used each type of health facility) sort the 
questionnaires into piles according to the first question (gender), then subdivide 
the piles according to the response to the other question (use of health facility). 
3. To do tally counting the basic procedure is:
o
One member of the compiling team reads out the information while the other 
records it in the form of a tally (e.g., /// representing 3 subjects, //// representing 
4 subjects who present a particular answer).
o
Tally count for no more than two variables at one time (e.g., sex plus type of 
facility used).
If it is necessary to obtain information on 3 variables (e.g., sex by time of 
attending a health facility by diagnosis), do a manual sorting for the first 
question, then tally count for the other two variables. 

---

Research methodology 
73
o
After tally counting, add the tallies and record the number of subjects in each 
group.
4. After doing either manual or tally counting, check the total number of 
subjects/responses in each question to make sure that there has been no omission or 
double count.
Note:
One can tally in two ways, 
. The latter way is preferable as it reduces the 
possibility of error. 
It should be noted that hand tallying is often used in combination with master sheet analysis 
when the relationship between two or three variables needs to be established, or details 
analyzed.