Qualitative Data Displays

It’s too bad that most talk about data concerns numerical data: survey responses and test scores (of various sorts). It’s only half the picture. Qualitative data comprise the other half.

Qualitative data tend, however, to be more unruly than numbers. We cannot just “add up” words or calculate an average.

Nevertheless, gathering data through the words of students, teachers, and parents can tell us a great deal about what they think, believe, or want. Their words can be data when teams treat them as data. Qualitative data come to teams in three basic ways: (1) as brief comments in response to open-ended items on surveys, (2) longer comments from interviews, and (3) notes from observations.

Although no one can use numerical procedures (like addition) on words, people have been summarizing verbal material for thousands of years. Teams can use various options for boiling down and displaying data that start out as words.

For some examples, let’s assume—as part of a long-term project related to fostering community engagement—a BLT has surveyed 250 parents to learn more about their opinions about the school and that the BLT has also conducted interviews with 25 of the parents. The team now has written responses to the surveys and transcriptions of the recorded interviews.

The survey included an item that asked:

The interview, consisted of five questions, one of which was:

The survey drew comments from 43 people who chose to respond to the open-ended item. In the interview, all 25 people provided comments in response to the question about how families would use community tutoring services.

How might a member of the BLT prepare these data to display them to the team?

The BLT member might just give the team all 43 comments from the survey. Of course, reading all of the comments might take too long and wouldn’t tell the team much about trends.

Alternately, the BLT member might group the responses into the ideas that were mentioned most often, and list those on a table such as the one below.

Clustering survey comments is good practice for summarizing interview material. Working with more extensive interview material—as with the transcripts produced in this example—can be time-consuming. Of course, if it’s the only way to answer certain kinds of questions, it’s probably worth the effort.

In this case, the interviews were used as a way to get a deeper understanding of what different families meant when they indicated their preferences for the various community tutoring services the school district might provide. So, data analysis would involve looking for patterns in the insights offered by parents. Organizing their insights into categories helps educators spot patterns. A display that looks for patterns based on family type (for instance, families with young children, families with children of various ages, families with mostly older children) might look something like the grid below. Color coding might help those analyzing the data identify possible patterns.

How would your family use community tutoring services if the school district could provide them?

Note that the color coding points to one likely pattern, namely that families of all types believe that tutoring services would be beneficial. Other possible patterns include the insight that various agencies also provide tutoring and that various types of peer tutoring might be helpful. The remaining insights also raise issues that relate to how students might access tutoring services—where the services are located, how students can get there, and when they’re provided. Like quantitative data, these qualitative data give educators a lot of information on which to base decisions about services that would help the district make better connections with families.