Digit preference = Sometimes people can find it difficult to recall certain details (say around aspects of their sex lives - which can span many years). So a person might provide their best guess in order to answer a research question. This can lead to ‘digit preference’ as the person is more likely to report a number ending in 0 or 5.

LGBT = This abbreviation stands of Lesbian, Gay, Bisexual and Transgender/trans. Sometimes the abbreviation includes other sex, gender and sexual minority groups, such as “Q” (for queer – a reclaimed word), “I” (for intersex – people born with diverse sex characteristics) and “A” (for asexual – people who are not sexually attracted to others). A “+” symbol is also frequently used to highlight that the abbreviation is intended to be inclusive of all those that are diverse in terms of their sex, gender and/or sexuality.

MSM (and WSW) = These abbreviations stand for “Men who have Sex with Men” (and “Women who have Sex with Women”). The focus here is on sexual behaviour  - not how a person identifies in terms of their sexuality (e.g., whether or not they are lesbian, gay or bisexual). Whether sexual behaviour, sexual identity or sexual attraction is the most important element of sexuality depends on the research question being asked. For example, some MSM will not identify as gay or bisexual, so public health campaigns and other interventions may focus on behaviour in addressing the sexual health needs of certain unique and under-served sub-populations, including MSM.  

Numerator = The numerator is the number ‘above the line’ in a basic fraction and it ‘sits’ above the denominator. For example, in Natsal-3 there were 8,869 female participants of the 15,162 participants in the study, so if working out the proportion of females in Natsal-3 then the numerator would be 8,869 (and the denominator would be 15,162). Expressed as a fraction that would be 8,869/15,162 (which works out as 58.5% of Natsal-3 participants are females).

Quantitative data = This describes the measures of values or counts as these are expressed in numbers. Natsal-3 consists of quantitative data, in essence numeric data about many variables (e.g. the how many and how often of certain sex and sexuality questions).

Denominator = The denominator is the total number of people a statistic is based upon. For example, in Natsal-3 there were 8,869 female participants of the  15,162 participants in the study, so the denominator here is 15,162. See ‘Numerator’ below for more details.

Recall bias = This is when there are differences in the accuracy (or completeness) of the recollections of participants regarding events or experiences from their past. For instance, they genuinely cannot remember how many sexual partners they have had.

Social desirability bias = There is a type of bias in research which is where a participant can tend to respond to questions in a manner that they think will be viewed favourably by others. For example, historically, men were more likely to ‘round up’ their number of sexual partners while women were more likely to ‘round down’ this number. This was because of cultural norms around what is seen as favourable for men and women in terms of the total number of sexual partners they ‘should’ have.    

Qualitative data = This describes qualities or characteristics. It is frequently collected via interviews and focus group and the audio-recorded conversations (i.e., the raw data) is transcribed. The data is often in the form of words and the words are examined for patterns or meaning, sometimes through the use of certain types of qualitative data coding.

Weighting = Weighting or statistical weighting involves emphasizing the contribution of particular aspects (e.g., groups) over others, frequently in order to ‘correct’ for an imbalance. For instance, in Natsal-3 certain underrepresented groups data is weighted slightly differently to ensure the results in analyses more accurately reflect what is estimated to be representative of Great Britain – in terms of sex and sexuality.

Participation bias = Participation bias (or non-response bias) is a phenomenon where the results of population-based research can become non-representative because certain participants disproportionately decline to take part, which impacts on the representativeness of the findings. For example, young men under 30 years old were most likely to refuse to take part in Natsal-3, meaning they are under-represented (although certain statistical tools are used to address this in analyses – in particular weighting).