Asking questions is widely perceived as the first step in science. It’s listed as the first NGSS Science and Engineering Practice, and it is, indeed, often how a scientific investigation formally begins. Yet something does happen before a question takes shape. Questioning also continues to happen after an initial question is posed.

Here are a few thoughts in hopes to breathe life into the practice of asking questions throughout the process of gathering, exploring, and making sense of data.

(Photo by MI PHAM on Unsplash)

Curiosity

Think of a question.

This is a hard thing to do when you don’t have a context, a situation, or background information that stimulates curiosity. Curiosity is a feeling that breeds questions naturally; if students don’t feel curious, questions will not come easily to them. One strategy to stimulate curiosity is to introduce students to situations, stories, or real problems that they are likely to become curious about. “What!?” “How can that be?” “What caused that?” “Is there a way to make it work better?”

Research questions

Once curiosity-questions start to flow, it becomes a matter of shaping them from the realm of curiosity into the realm of finding out. Not all curiosity-questions need to make it to the ‘finding out’ stage. Students can learn to discern questions that can be feasibly investigated and those that can’t. They can learn to reframe curiosity-questions to something more constrained that can be researched. For example: Why are some people bitten by mosquitoes more than others? How have humans affected Sebasticook Lake? How can it work better? are rich curiosity questions, but they are not easy to investigate in their current form. Such questions can be reframed to something more constrained that probes part of the original wonderment: Are mosquitoes more attracted to different colors? How has water clarity changed in Sebasticook Lake over the past 20 years? Which blade design makes our windmill spin faster?

Research questions can be worded in a way that gives clues about how to proceed with an investigation. For example, the question about blade design suggests what the experiment should involve (i.e. two or more different blade designs) and what needs to be measured (i.e. how fast they spin, such as number of spins per second or per minute). It is a question about comparing numeric measurements made for two or more groups. The question about water clarity is about something changing over time. It also suggests what measurements are needed (water clarity values for Sebasticook Lake) and a time range (the past twenty years). Note that the question does not suggest at what time-scale the measures should be made, whether weekly measurements, or monthly, or yearly averages. This detail is left up to the investigator, and may depend upon what data are available. 

In this way, a well-framed research question does more than just pose a researchable unknown, it suggests what kind of study is needed.

Meta-questions

Once students have questions that they can feasibly investigate, they will be faced with the big question of how to proceed. (Hopefully they will be encouraged to face that question.) Unless they are following a cook-book lab with step by step instructions, they will need to ask questions such as What should we measure? How often should we measure it? How should we organize the data we collect? Once they have their data, they’ll need to ask Which parts of the data do we need to answer our research question? What type of graph should we use? What inferences can we make? How sure can we be? These are meta-questions — questions about the data and the process of sense-making.  

Meta-, from Greek, has several meanings; perhaps the most relevant here are “after, behind, among, between“. Meta-questions are questions behind the questions. Do we have enough data? Could there be bias in our sample? Too often, students have been excluded from asking meta-questions; lab books provide instructions for what to do, tables for filling in data, and even a grid set-up with instructions for graphing results. Do these values make sense? How else could we look at the data?

Questions as engines

If data are the fuel, questions are the engine that transforms the fuel into useful information. Questions — curiosity questions, research questions, meta-questions, all kinds of questions — move the process forward throughout the effort.

But, as with an engine, a spark needs to start it. The spark is curiosity. It’s just a matter of helping the engine hum along after that.