Fieldwork for the third survey, Natsal-3, involved interviews with 15,162 adults aged 16-74 years between September 2010 and August 2012.

Natsal-3 comprised six interlinked work packages (WPs). These were:

WP1: Development phase

WP2: Core National Survey of Sexual Attitudes & Lifestyles

WP3: In-depth qualitative exploration of behaviours

WP4: Population-based measures of Sexually Transmitted Infections

WP5: Population-based measures of sex hormones

WP6: Web survey comparison study

The core survey is a population-based probability sample survey of 15,000 men and women aged 16-74 years, resident in Britain (England, Wales and Scotland), which measures a wide range of variables related to sexual attitudes and lifestyles.


Development work

Due to the challenges of collecting accurate data on sexual behaviour and the increased scope of the third survey, there was an extensive two year development phase for Natsal-3. During this period, the research team developed new questions within pre-identified areas of enquiry. Key questions (new and existing) were cognitively tested to ensure they were comprehensible and acceptable to the older age group, included for the first time in Natsal-3. Psychometrically validated measures of sexual function, and of unplanned pregnancy, have also been developed as part of this development work.

The first pilot of Natsal-3 took place in August 2009. In the light of the pilot findings, the research team altered and improved fieldwork procedures before the second pilot which took place in March 2010.

Qualitative work

An innovation for Natsal-3, this work package followed-up selected survey respondents with in-depth interviews to explore sexual practices and experiences in more depth. The specific objectives of this work package were to:

  • further illuminate associations between variables found in Natsal-3 data;
  • examine the context, meaning and significance of specific experiences to individuals;
  • understand factors enhancing and hindering the improvement of sexual health status.

The focus of this work package was on sexual health outcomes of scientific and policy interest. We talked to people who are difficult to sample by other means, about topics on which there was little existing qualitative data. The sampling frame comprised participants in Natsal-3 who expressed willingness to be re-contacted to take part in further research and who fall into at least one of the following five groups:

  1. Women aged under 35 who had conceived a child before age 18
  2. Men who paid for sex in the last five years
  3. Men and women aged over 40 years having unsafe sex (multiple partners and inconsistent condom use)
  4. Men and women aged 45-55 years reporting lack of satisfaction with their sex lives
  5. Men and women aged 55 years or older reporting sexual difficulties and ill health

The combination of quantitative and qualitative data within a common sample provides opportunities for both descriptive and interpretive analysis. The results informed the design of preventive interventions, including those involving sexual health services, sex education, and sexual health communication campaigns.

STI Measures

Knowledge of the epidemiology of STIs in the UK is largely based on cases reported by genito-urinary medicine (GUM) to the Health Protection Agency (HPA). Diagnoses have been increasing over the last decade. However, routine surveillance gives only part of the picture of STI epidemiology. It does not measure the true extent of STIs in the population or the prevalence of asymptomatic infections and only limited socio-demographic and behavioural information is collected.

We have collected urine samples from a proportion of Natsal-3 respondents. These samples will be tested for five infections that may be transmitted sexually: Chlamydia trachomatis, Neisseria gonorrhoeae, mycoplasma genitalium, human papillomavirus and HIV. We have collected urine samples from around 5,000 men and women aged 16-44 years and remaining urine is being stored for future testing.

Analysis of these samples will allow us to:

  • measure the population prevalence of these five STIs
  • identify demographic and behavioural factors and clinical symptoms associated with each infection

Testing of urine samples in Natsal-2 for Chlamydia trachomatis informed the strategy of the National Chlamydia Screening Programme, which initially focussed on women aged 16-25 years, but was then extended to include young men.

Hormone measures

As part of the development phase for Natsal-3, we validated an assay to measure levels of testosterone in saliva. Salivary testosterone is related to the concentration of 'free' testosterone in serum; the component available to tissues that correlates better with androgenic status. The assay was validated against levels in blood, which is the usual way of measuring testosterone. Remaining saliva will be stored for future testing.

We collected saliva samples from approximately 4,500 men and women aged 18-74 years.

Analysis of these samples allowed us to:

  • explore the relationship between free testosterone levels in saliva and sexual behaviour and function;
  • establish the normal range of free testosterone across the life span in men and women; and
  • document the prevalence of late-onset hypogonadism (testosterone deficiency) in middle aged and elderly men
Web survey comparison study

In 2011, the Wellcome Trust provided funding for additional methodological work to develop a shorter version of the Natsal-3 questionnaire to be administered as a web survey. Three well-known market research organisations in the UK were commissioned to carry out four surveys among their existing web panel members, and their results were compared with Natsal-3 and, for a few socio-demographic variables, with external benchmarks. Two of the web surveys set quotas on 'basic' variables (sex, age, and partnership status), while the other two were 'modified' to set quotas on additional variables known to be related to some of the key behavioural measures.  These surveys were restricted to panel members aged 18-44 years. The main objectives of this work were:

a) to examine the extent of agreement between results from the surveys carried out among volunteer web panel members with each other and with those obtained from the Natsal-3 CAPI/CASI probability sample survey and

b) to determine whether modifying the quota controls would lead to web panel results more similar to those found on Natsal-3.

Another component of the work involved following-up a sub-set of participants in the main Natsal-3 interview survey. About one to two months after their CAPI/CASI interview, 527 Natsal-3 participants completed the web version of the questionnaire. The aim of this part of the project was to give an indication of the willingness of participants to complete the survey on the web and to provide direct comparison of responses to the same questions asked in CAPI/CASI and on the web.

A final part of the project involved one of the market research organisations obtaining samples of 500 participants of two minority groups (gay men and Black Africans) who were represented in small numbers even in large surveys like Natsal-3, and are difficult (and costly) to 'boost' using probability sampling methods.

Contact us