Instructions: Findings - Summarize Data
DISCLAIMER: This data in this section is fictitious and does not, in any way, represent any of the programs at Gallaudet University. This information is intended only as examples.
Information is the results of data processing, or said another way, data only becomes information that you can use to make decisions after it has been processed.
It's hard to understand data in bulk. Thus, it's best if the data is summarized in the results.
The benefit of summarization is that it not only reduces the amount of data needed to digest, but it increases the ability to interpret the data.
Tips to Summarize Data
Organize the Data
If there is a small amount of data, it can be prepared it by hand. Otherwise, the results should be entered into a computer for easier summarizing and analyzing.
If the assessment tool uses descriptive instead of numeric categories, it will be necessary to change the ratings or responses into numbers (coding) before entering them into the computer. It will make them easier to summarize and analyze.
Exemplary = 4
Can express why psychology is a science = C1
Notes on coding. Keep careful notes explaining the meaning of each code to minimize confusion. They will be invaluable if anyone decides to repeat the assessment later.
Summarize the Data
Clean the Data
Depending on the data collection, a cleaning up will be needed to make sure it is appropriate and accurate prior to being summarized and analyzed. For example, assessment results from a paper-based survey or rubric may include some unclear or inaccurate responses that you will need to be decided about (e.g., correcting or eliminating data from the sample).
Some types of responses that may need to be address before summarizing data:
- Inapplicable responses
(e.g., males students answered questions in section for female students only)
- Inappropriate multiple responses
(e.g., two answers checked for one non-multiple choice question)
- Responses outside given category:
(e.g., student wrote in answer because they didn't like choices provided)
- "Other" responses that really aren't
(i.e., student checked “Other — Please Specify” but their comment matched one of the answers provided)
- Make a List (& check it twice)
- List the raw data
- Remove identifying information such as names to ensure confidentiality
- Compare the list to the source information. This will help in finding and correcting any errors.
Once the list is accurate, proceed to the next step.
4= Exemplary 3 = Good 2 = Minimally Acceptable 1 = Unacceptable
4. Tally the results or responses to get a quick picture
Example: Tally of raw data from list above
- Chart Your Results in a way that is meaningful. It is often helpful to use tables, line graphs or bar charts to get a clear look at the big picture. It depends on the kind of questions the assessments are needed to answer. (see two examples below showing the same data summarized two ways).
- AVOID complex statistics
- Use round numbers
- Create simple charts, graphs, lists (They are easier to read and understand.)
- Sort results from highest to lowest [optional]
- Percentages may be more meaningful than averages
- Show trend data if assessing over time
Example 1: Table using data from tally above with percentages added, column with total percentage of students who were successful in the program (adding Exemplary + Good + Minimally Acceptable)
Example 2: Line chart using data from tally above with target the program hope to achieve.
Find the Story in the Data [Analyze Data]
Data summaries make it easier for you to see meaning but by themselves they don't reveal the whole story. You also need to include an explicit narrative interpretation of what you saw in the data…and what you plan to do about it.
- What do the data summaries reveal about students' learning? (identify meaningful information)
- What are you going to do about what you have learned?
- When, where, and how are you going to do it?
Additional Resource: More examples of summarized data are in the attached document, including a thematic analysis of qualitative data.