Crafting SMART Research Questions

The Art of Reformulation

Reformulation is not about finding one "correct" question, but about making deliberate, defensible choices to address the weaknesses identified during your critique. It's a constructive process where you define your population, select your variables, and set the scope of your inquiry.

To guide this process, let's explore a resource that outlines the principles of crafting a strong research question and provides clear before-and-after examples.

As the resource highlights, the key is to move from the broad and subjective to the narrow, specific, and measurable. Let's apply this principle to our examples.

Case Study 1: Reformulating a Bra Design Question

In our last lesson, we critiqued the following question and found it failed on all five SMART criteria.

Flawed Question: "To determine if softer materials make bras more comfortable."

Let's rebuild it, addressing each criterion.

• S (Specific): We must define our terms.

  • Population: Instead of all women, let's focus on a specific market segment with known comfort challenges: women wearing sizes 34DD to 38G.
  • Intervention: Instead of "softer materials," let's compare a specific new fabric, a modal-spandex blend, against the current standard, nylon.
  • Context: We'll test this in a specific product: the 'Aura' full-coverage underwire bra model.

• M (Measurable): We must operationalize "comfort." As a statistician, you know that a latent concept like comfort must be represented by measurable variables.

  • Metric 1 (Subjective): A self-reported comfort score on a 10-point Likert scale, administered at the end of an 8-hour wear period.
  • Metric 2 (Behavioral): The number of physical adjustments (e.g., strap pulling, band adjusting) made by the participant during the final hour of wear, as counted from a video recording.

• A (Achievable): A study testing all materials is impossible. Our refined scope is achievable.

  • Plan: We can recruit a sample of 50 women who meet the size criteria, provide them with both versions of the 'Aura' bra, and have them complete the wear test on two separate days. This is a feasible project for a small R&D team.

• R (Relevant): The question must link to a clear business or research objective.

  • Goal: The research is relevant to the broader goal of reducing product return rates attributed to poor comfort from 15% to under 10% for the upcoming product launch.

• T (Time-Bound): We need a concrete schedule.

  • Timeline: The wear tests will be conducted over a 4-week period in Q3, with data analysis and a final report due by the end of October.

The Reformulated SMART Question:

For women wearing bra sizes 34DD-38G, does the 'Aura' underwire bra made with a modal-spandex blend achieve a statistically significant higher mean comfort score (on a 10-point scale) and a lower average number of physical adjustments, compared to the same model made with standard nylon, based on an 8-hour wear test conducted over a 4-week period, to inform a material choice aimed at reducing customer return rates?

This reformulated question is now a complete research plan in a single sentence. It tells us who to study, what to do, how to measure the outcome, why it matters, and when it will be done.

Case Study 2: Reformulating a Relationship Dynamics Question

Next, let's apply the same constructive process to the social research question we critiqued.

Flawed Question: "Investigate how polyamorous people manage their schedules."

This question was also too broad and vague to be researchable.

• S (Specific): We need to define both the people and the activity.

  • Population: Let's narrow from "polyamorous people" to a specific structure you are interested in: one-female-multiple-male (Vee or Triad) polyamorous relationships where all partners have been together for at least one year.
  • Activity: Instead of "manage their schedules," let's focus on a specific, often challenging aspect: the process for negotiating and allocating one-on-one time between the female partner and each of her male partners.

• M (Measurable): We need to quantify the "how."

  • Metric 1 (Process): The primary variable of interest will be the scheduling strategy used, categorized as 'formal' (e.g., using a shared calendar, weekly meetings) or 'informal' (e.g., ad-hoc requests).
  • Metric 2 (Outcome): We will measure relationship satisfaction using a validated instrument like the Relationship Assessment Scale (RAS), administered to all members of the relationship.

• A (Achievable): A study of "all polyamorous people" is not feasible. Our focused approach is.

  • Plan: We can recruit a sample of 15 to 20 such Vees/Triads through specialized online forums and community groups for a qualitative and small-N quantitative study. This is a realistic scope for a graduate-level research project.

• R (Relevant): The research should serve a purpose beyond simple description.

  • Goal: The study is relevant to the goal of identifying scheduling strategies that correlate with higher relationship satisfaction, in order to develop educational resources for the polyamory community.

• T (Time-Bound): We need a defined data collection period.

  • Timeline: Data collection, consisting of an initial demographic survey and bi-weekly satisfaction surveys, will take place over a three-month period.

The Reformulated SMART Question:

Among one-female-multiple-male polyamorous relationships of at least one year's duration, is there a significant difference in average relationship satisfaction scores (measured by the RAS) between those who use formal scheduling strategies (e.g., shared calendars) versus those who use informal strategies for allocating one-on-one time, based on data collected over a three-month period, to identify practices that support relationship health?

Once again, we have transformed a vague interest into a precise, answerable, and compelling research question.

Your Turn to Reformulate

Now it's your turn to practice. In the previous lesson, we critiqued the following question. Take a moment to reformulate it into a full SMART question, applying the principles we've discussed. Be specific about the population, the measurement, and the relevance.

Flawed Question: "To understand jealousy in polycules."

After you've drafted your own version, compare it with the example solution below.

Example Solution:

Here is one possible SMART reformulation. Your version may differ in its specifics, but it should exhibit the same level of detail across the five criteria.

• S: Investigating feelings of sexual jealousy experienced by the pre-existing female partner in a one-female-multiple-male triad, following the introduction of a new male partner.
• M: Jealousy intensity will be measured weekly using the Multidimensional Jealousy Scale (MJS).
• A: A sample of 20 triads that meet this specific criteria will be recruited through online support groups.
• R: The research is relevant to developing therapeutic interventions to help established partners navigate the emotional challenges of polycule expansion.
• T: The study will track jealousy levels for the first six months after the new partner joins the triad.

Synthesized SMART Question:

In one-female-multiple-male polyamorous triads, how does the intensity of sexual jealousy (measured weekly by the MJS) in the pre-existing female partner change over the first six months following the introduction of a new male partner, to inform the development of targeted therapeutic support strategies?

Conclusion

This lesson concludes our deep dive into the SMART framework. By moving from critique to reformulation, you have mastered the complete process of refining a research idea into a viable plan.

Key Takeaways:

• Reformulating a research question is a constructive process of making specific choices to address vagueness, imprecision, and lack of focus.
• A fully reformulated SMART question acts as a blueprint for your research, defining the who, what, how, why, and when of your study.
• This skill is fundamental for any researcher, ensuring that the data you collect will be capable of answering the question you set out to investigate.

Preview of the Next Module:

Now that you can formulate a powerful and precise research question, our next step is to consider how we will go about answering it ethically. In Module 2, "Ensuring Data Ethics with CTO POC," we will explore the critical principles of Consent, Transparency, Openness, Privacy, Ownership, and Currency. Asking the right question is the first step; asking it in the right way is the next.