| You have reviewed the data
and formed a general sense of what you have at your
disposal. Now, it’s time to get focused and organized.
The data need to be sorted, categorized, and organized
in a manner which will allow you to understand and communicate
the results. In most cases, you will prepare and organize
your data with the help of a computer, and possibly
additional staff. Either way, there are several things
that should happen to prepare your data. |
|
To make the great volume of data you have collected
manageable, follow these steps:
- Reduce, condense, or transform the data.
This means eliminating inappropriate or meaningless
items from the data group, putting items together
(e.g., adding up knowledge items to get a knowledge),
and turning the data into information that answers
the questions being addressed.
Data reduction often forces you to make choices about
which aspects of the assembled data you want to emphasize,
minimize, or set aside completely for the purposes
of the project at hand.1
Some examples of data that could be eliminated
include2:
-
Uninterpretable responses like
“Does not apply”, “Don’t
Know”, “No response” or other
unusable or illegible responses
-
Responses that are insincere
or unlikely. (e.g., Question: Do you have any
suggestions for improving the program? Response:
“Give every participant a million dollars.”)
-
Items for which respondents
have selected more than one answer (e.g., “strongly
agree” and “agree”)
-
Items where the respondent has
chosen the same answer for every multiple choice
question (e.g., “strongly disagree”
with every statement)
-
Reporting allocations of time
(or other parts of a whole) that add up to more
than 100 percent (e.g., 40% of the program was
devoted to education, 40% was devoted to role-playing,
and 40% was devoted to discussion)
-
Inconsistent answers (e.g.,
Question: Have you ever smoked cigarettes? Response:
“No.” Question: How long ago did
you quit smoking? Response: “More than
2 years.”)
-
Responses that clearly indicate
a misunderstanding of the question (e.g., Question:
Where did you learn about the dangers of second-hand
smoke? Response: “Yes”)
- Categorize or classify the data
so that they can be easily processed. This is mainly
required for items that are not pre-coded, for “open-ended”
answers, and comments by respondents and observers.
Sometimes, however, it is necessary to combine infrequent
responses into a category of “other”.
- If using a computer, choose a software package
that is easy to use, offers you a framework for data
entry that works with the information you have, and
allows the kind of analysis and presentation you want.
- Create a database to house and
organize the information. This tool can vary from
a simple spreadsheet completed by hand, to a sophisticated
computer-based tool using powerful off-the-shelf database
software.
- Enter the data into the database.
Depending on how much data you have and how political
and sensitive your project and its data are, you might
want to institute quality control procedures for coding
your data. This requires assigning people to check
and double check the work being done, and verify the
data entered.
- Conduct a final check (or “data
cleaning”) to look for coding and entry errors.
Chances are, the data will be error free if the coding
and entry were carefully completed the first time
around.
- Make extra copies of your data, and keep
the master in a safe location. Use a copy
for making any changes, cutting and pasting, creating
new variables, etc. Save this copy as the master,
only when you are sure that changes have been made
successfully.
Return to Analysis Plan
---------------
1. Source: Frechtling, J., Sharp, L.,
& Westat, eds. (1997). Analyzing qualitative data.
In: User-friendly handbook for mixed method evaluations.
Arlington, VA: National Science Foundation, Directorate
for Education and Human Resource. web site.
2. Source: Action & Research Open
Web. (n.d.). Data analysis. In: Module 6: Project
Evaluation. Retrieved July 21, 2004 from the University
of Sydney’s web site.
Bibliography
Source: McNamara, C. (1999). Analyzing,
interpreting and reporting basic research results.
Retrieved July 21, 2004 from The Management Assistance
Program for Nonprofits web site:
http://www.mapnp.org/library/research/analyze.htm
Source: Stevens, F., Lawrenz, F., &
Sharp, L. (n.d.). Design, data collection and data analysis.
In: User-friendly handbook for project evaluation:
Science, mathematics, engineering and technology education
(pp. 31-58). Arlington, VA: National Science Foundation,
Directorate for Education and Human Resources.
|
