Designing questions for research

The reading age of a question should be below the youngest participants in the study. Tools such as this readability test can assist. Use specific, straightforward language and ensure that only one concept is included in the question. For example, in the item ‘I sometimes feel pressure to do sexual activities I am not comfortable with’, participants may be unsure what is included by ‘sexual activities’. Adding brackets with clear examples (e.g. kissing, cuddling or having sex) makes it more specific. 

It cannot be assumed that participants have shared understandings of terms, or that they understand them at all. For instance, the term ‘sex’ may mean penis-in-vagina sex to most participants but may mean oral and/or anal sex to others. Some terms that may appear basic – like ‘heterosexual’ and ‘vulva’ – are often poorly understood by young people. Definitions for essential terms like these should be provided proximate to, or within, the question, or via hyperlink to a glossary.

Language related to sex evolves continually; new terms emerge, and existing ones may fall out of currency. For example,  the term ‘homosexual’ is less common than previously. Words and terms can also shift in meaning across time and context. For instance, many LGBTQ groups use ‘queer’ as an umbrella term, reclaiming it from its previous derogatory use. Many terms in everyday use are too vague and euphemistic to be useful in research. This includes terms like ‘speaking to’ or ‘getting off with’ to describe an early phase of a relationship. Terms like boyfriend/girlfriend can be problematic in that they suggest a gender binary, but a term like ‘partner’ is not always well understood by young people.

Being inclusive means ensuring that questions and their response options capture the full range of possible experiences. Specifically for this topic, it means avoiding language that assumes young people are heterosexual and cisgender. Small adjustments to phrasing can turn an exclusionary sentence into an inclusive one. As an example, the singular ‘they’ could be used instead of ‘she’ and ‘he’ to include all gender identities. However, being inclusive should not come at the expense of clarity to participants. For instance, if use of the singular ‘they’ makes the question unclear then a different solution to inclusivity must be found.

Embedded assumptions presuppose information about the participant. An example is a question asking whether contraception was used at last sex – which assumes that last sex involved a penis in a vagina (i.e., a risk of pregnancy).

Embedded assumptions can result in leading questions and response bias. 

Adolescence is a period of rapid and often dramatic shifts in physical growth, knowledge and experience. Age is a proxy for this transition but there is also significant variation at each age. For example younger adolescents may have begun thinking about sexual attraction but not yet about sexual identity. Careful use of filtering (e.g., skip patterns) can assist in directing participants to relevant questions. Understanding what is appropriate to ask when, should be based on existing evidence and the views of young people.

Change over time in how young people feel and identify is common and there is not always a strong overlap between reported attraction, identity and behaviour at any point in time. Allowing ‘unsure’ and ‘other – please describe’ options can give participants opportunity to express uncertainty they may feel about their developing identity and feelings. Surveys that only ask about behaviour or identity or attraction should be aware that each captures a different population.

For more information on how to ask about attraction, identity and behaviour, see Researching and monitoring adolescence and sexual orientation: Asking the right questions, at the right time.

To learn more about the implications of different question types for capturing sexual and gender minoritised populations see Sexual Orientation and Gender Identity/Expression in Adolescent Research: Two Decades in Review. 

Allow young people to use language with which they feel most comfortable, except for offensive slurs or insults.

Group interviews with younger adolescents may involve embarrassment that is masked by provocative or challenging language. Facilitators may need to gently but directly confront this and ask the group to agree on acceptable terms. If young people are unsure whether a word is offensive, encourage open discussion about this and use these conversations to convey appropriate/accurate terms (for example, when referring to body parts or sexual acts).

Some words used previously (such as queer) have been reclaimed and have new and positive meaning so it is advisable to check their understanding and use.