EN FR

The research question

DSN201 - 3. The research question

The research question

The research question is a pivotal element in the development of a research project, between an observation of the study context and a directed analysis that will attempt to propose original knowledge. It is often the logical follow-up to the elucidation of the study context (we find ourselves in a given situation, with specific stakes and unknowns, therefore our research question is as such). It precedes the statement of the research conduct (which, of course, is based on the research question).

The research question is therefore the driving force behind the research conduct: it justifies the resources committed and guides the research. Coupled with the research hypothesis, which normally follows it, it is the foundation of the research and of the scientific article.

Relevance of the research question

The relevance of a research question manifests through the understanding of the context in which it is formalised, through its originality (i.e., the fact that the answer is still unknown), through its timeliness, through its operability (it can be answered scientifically) and through its clarity (one will be able to clearly determine whether the question has been resolved or not). It is all the more appreciated if it offers a new perspective on a topic debated within the research community.

In the field of medicine, certain criteria are described for developing a good research question (adapted here). The research question must be:

  • Feasible — It must take into account the limits and problems related to the research work:

    • Access to necessary resources — one must be able to obtain sufficient data (e.g., quantity and quality of subjects interviewed, access to literature on the topic),
    • Sufficient technical expertise — one must be able to handle the tools or other means necessary to conduct the experimentation: skills, equipment, experience to design the experimentation, recruit subjects, measure variables, conduct the analysis and discuss the results. It is often recommended to use known and recognised techniques and methods, rather than developing new ones oneself,
    • Time and financial resources — one must have the time and resources to carry out the research. An ethics committee, for example, can be very time-consuming and compromise the research; the need to travel, use expensive equipment… can also be a difficulty that jeopardises the research activity,
    • Restricted question — a question that is too open can deteriorate the quality of the research. It is very often preferable to develop a restricted question, which will improve its effectiveness.

  • Interesting — If you are the only person interested in the question, that is a bad sign…

    • The answer must interest both colleagues in the research field (which validates the possibility that this research participates in “the scientific discussion”) and other people (which validates a probably practical interest beyond the research itself). Verifying this interest involves constructing a research narrative, which will allow validation of the clarity and potential interest of the research, as well as increased visibility of the project,
    • The interest may come from the fact that the research attempts to confirm, refute or build upon previous results, or on the contrary that it attempts to propose new knowledge, which will need to be justified during the development of the research question or hypothesis,
    • A personal interest can also be put forward (e.g., visibility to contribute to a call for funded research projects).

  • Ethical — The ethical dimension of research primarily requires approval from an institutional ethics committee, which judges the respect and protection of people engaged in the research. The difficulty of the process and its timeline can be a risk for the execution of the research.

    In design research, ethics takes on specific dimensions:

    • Informed consent for prototypes — Participants must understand that they are interacting with a research prototype, not a finished product. Consent must cover the use of data generated by the interaction.
    • Privacy in contextual research — In situ studies (ethnography, probes, observations) often collect sensitive data about living contexts. Anonymisation and protection of this data are crucial.
    • Responsibility of the researcher-designer — The designer-researcher assumes responsibility for the artefacts they produce, particularly when deployed in real-world contexts. The social, environmental and political implications of designs must be anticipated.
    • Vulnerability of participants — Some research involves vulnerable populations (children, elderly people, marginalised communities). Additional precautions are necessary, and the ethics committee may require enhanced safeguards.

    Sengers, Boehner and Warner (2015) propose a reflective HCI approach that invites the researcher to question their own values and those they embed in their designs. Le Dantec, Poole and Wyche (2009) show how values emerge from users’ lived experience and must be discovered rather than presupposed.

    References

    • Le Dantec, C., Poole, E. S., & Wyche, S. (2009). Values as lived experience: Evolving value sensitive design in support of value discovery. Proceedings of the 27th International Conference on Human Factors in Computing Systems, 1141–1150. https://doi.org/10.1145/1518701.1518874
    • Sengers, P., Boehner, K., & Warner, D. (2015). Reflective HCI. Proceedings of the 2015 ACM SIGCHI Conference on Human Factors in Computing Systems, 3749–3752. https://doi.org/10.1145/2702123.2702491

  • Impact — Relevance can concern several aspects, and most often boils down to asking the question of the impact of the research results on:

    • the discussion within the scientific community,
    • scientific knowledge,
    • the social or industrial situation,
    • future research questions.

The research question therefore requires good knowledge of the recent state of the art around the topic it addresses (to avoid trying to answer a question that has already been resolved). The literature review is essential. A discussion with recognised experts on the topic can also be beneficial.

New techniques, new tools, new study contexts… are opportunities to question topics from new angles, and thus in an original and timely way in relation to newly accessible resources.

The form of the research question

In its form, it can of course be written directly or indirectly. Example based on (Deckers et al., 2013), combining the research question and hypothesis:

  • Direct version: how do the experiences of people and the interaction between person and artefact change comparing explorative, active and following behaviours [of the artefact]. Following the theoretical outline related to human-human interaction, we hypothesise that the experience of the perceptual crossing [caused by the explorative behaviour] between person and artefact will positively impact the appreciation in interaction.
  • Indirect version (original): In this experiment, the explorative behaviour is compared to the active and the following behaviour to see what the effect is on the experience of people and the interaction between person and artefact. When perceptual crossing between person and artefact happens, this will, following the theoretical outline, cause the person to feel involved in the situation. You will come to appreciate that you affect and are affected by the artefact. An artefact and a person can build a common history in the course of their interaction.

The question, the hypothesis and the strategy

A research question is most often accompanied by a research hypothesis, which the research question attempts to answer, and a research strategy, which will make it possible to answer it. The feasibility and clarity of this research strategy often demonstrate the capacity and effectiveness of the research to provide a satisfactory answer.

The development of the research strategy must be done rigorously, to anticipate risks and potential difficulties upstream, avoid them if possible or prepare for them otherwise.

The research question, the research hypothesis and the research strategy are the foundations of practice; the quality of the research execution depends on their development. It is therefore advisable to be continually critical during their development, and to present these elements to colleagues, to receive recommendations for improvement before it is too late.

The research hypothesis

A hypothesis is a verifiable proposition or supposition about a relationship between two or more variables. It is built from existing knowledge, established by the state of the art, and from observations. Research most often consists of testing the hypothesis through experimentation (i.e., empirically).

Often, two hypotheses are confronted:

  • the null hypothesis assumes that there is no relationship or effect between the variables,
  • the alternative hypothesis assumes that there is a relationship or effect.

The quality of a hypothesis is evaluated by:

  • its clarity, precision and concision, as well as those of its variables,
  • its proper contextualisation,
  • the possibility of testing it: evaluable and falsifiable (it must be possible to attempt to prove it wrong).

Evaluating the research hypothesis

A hypothesis is most often evaluated according to a five-step process:

  1. Study design
    • Choose a type of research (quantitative, qualitative or mixed),
    • Define the sample, the data collection tools and operationalise the variables.
  2. Data collection
    • Collect data through surveys, experiments, interviews or observations,
    • Ensure the reliability and validity of the instruments used.
  3. Data analysis
    • Use appropriate statistical tests (t-test, ANOVA, regression, etc.) for quantitative data,
    • For qualitative data, use methods such as thematic analysis or content analysis.
  4. Interpretation of results
    • If the results reject the null hypothesis and support the alternative hypothesis, the latter is validated,
    • Otherwise, the null hypothesis is retained, without however proving it: one can only say that it has not been rejected.
  5. Presentation of results
    • Include indicators such as statistical significance, effect size and confidence intervals,
    • Discuss the implications, limitations of the study and future research directions.

The case of the physical hypothesis

Echoing research through design (see 2. La recherche: prototypes are described there as “material hypotheses”), the prototype often takes the role:

  • of physical hypothesis (physical hypothesis with sufficient product qualities — Frens, 2006, p. 185),
  • of embodied hypothesis (embodied hypothesis with sufficient system qualities — Hengeveld, 2011, p. 200),
  • or of behavioural hypothesis (behavioural hypothesis with sufficient behavioural qualities — Deckers et al., 2013, p. 28).

Some references

  • Deckers, E. J. L. (2013). Perceptive ‘qualities in system of interactive products’ (Doctoral dissertation, Eindhoven University of Technology).
  • Frens, J. W. (2006). Designing for rich interaction: Integrating form, interaction, and function (Doctoral dissertation, Eindhoven University of Technology).
  • Hengeveld, B. (2011). Designing LinguaBytes: A tangible language learning system for non- or hardly speaking toddlers (Doctoral dissertation, Eindhoven University of Technology). https://doi.org/10.6100/ir715535