Over the last few years there have been major improvements in sex research. The methods can differ on the following:
The measurement of sexuality: this can be self-report, biological measures or observation.
The number of subjects studied, either large or small groups.
Whether studies take place in the field or in a laboratory.
Sex can be manipulated or studied in natural occurrence.
It is important to study and understand these techniques and their limitations. It gives a better understanding in current sex research, but also in the research that is yet to come.
The first thing that needs to be decided is how to measure sexuality. There are several methods available:
Self-reports are the most common way and exist of questions about people’s sexuality. They can be conducted on paper, with interviews or online.
Behavioural measures include direct observation where participants are observed, and the behaviour is described. Another behavioural measure is eye-tracking where participants’ eye movements are studied when they look at images on a computer. Illegal sexual behaviour is studies using police reports. In the FBI they use uniform crime reports to study illegal sexuality. However, it only detects cases that are reported to the police.
Implicit measures include the implicit associations test in which the strength of associations to sexual content is measured. People tend to respond faster when an association is stronger. People are asked to divide sexual images or text into categories, such as pleasant and unpleasant.
Biological measures are measures of response to sex. There are genital measures in which arousal is measured by vaginal changes in women and erection changes in men. MRI and fMRI are now increasingly being used, they look at autonomy and brain regions involved in sexual behaviour. Strengths are its normativity, but limitations are the complicated statistical research needed to read the results and there is a lot of noise if the participant does not lie completely still in the scanner. Also, the results depend on the choice of sets of stimuli.
One important consideration is the population that is being used. The question here is who do we want to study? A scientist should always take a sample of the population. If the sample is a random sample, the results can be generalised to the population that was originally identified. However, random sampling is difficult because selecting a random one out of every fifty people in the United States can be tricky. Therefore, probability sampling is often used. Here, participants are selected based on characteristics meaning that the results are not generalizable to the whole population. Sampling exists from three stages:
Identification of the population.
Choosing a method for sampling.
Contacting the people in the sample. Here, researchers encounter the problem of refusal or non-response. The researcher needs to use volunteers, which causes the sample to contain volunteer bias, people that participate might be different from the ones that refused to participate. Moreover, women are less likely to volunteer for sex research than men.
Because there are some issues with sampling, convenience sampling can be used. It includes volunteers that came to sex’ therapist offices and volunteer to participate. It is chosen in a haphazard manner and relative to the population of interest. It differs from a probability or random sample.
Another issue is the accuracy of measurement. Participants in sex research often engage in purposeful distortion, when giving self-reports, they distort reality. They might exaggerate sexual behaviour (enlargement) or conceal certain things (concealment). To avoid distortion the participants are told that for scientific purposes, the information they give must be as accurate as possible, besides that anonymity must be guaranteed. However, even with these measures, there are two factors that impact the accuracy of responses:
Memory recall is needed to retrieve information on sexual behaviours years ago. One solution is asking about current sexual behaviour. However, children cannot be asked due to ethical and practical problems. One alternative is to use daily diaries in which people report their behaviour.
Difficulties with estimates arise when people are asked about frequencies. People cannot estimate very accurately which leads to inaccuracy in the self-report data.
Evidence of the reliability of self-reports can be measured. One measure is test-retest reliability where the respondent is asked to answer a series of questions once, and again when some time has passed. The correlation between the answers is noted as test-retest reliability. Another option is using a computer-assisted self-interview (CASI) method. The questions can be read aloud in addition to written questions for people that are poorly in reading questions, while giving the same level of privacy.
Accuracy of behavioural observations is an issue with natural observation of sexual behaviour. Direct observations have an advantage on self-report measures in accuracy but have their own problems. For example, they are expensive, time consuming and only a small sample can be studied. The sample that can be used probably consists of an unusual population that allows researchers into their bedroom.
Extraneous factors can influence sex research, these factors include race, gender or age of the respondents and the researchers.
Ethical issues arise when participants feel their privacy is being violated. A solution can be the use of an informed consent, where participants have the right to be told the purpose of the research. Another solution is the protection from harm where the levels of stress and discomfort are minimised. The principle of anonymity is an example of protecting participants from harm. The justice principle means that risks and benefits of participating should be equally distributed across research groups. A cost-benefit approach means that the stress of participants should be minimised and that what is left is the cost. Researchers should consider if the benefits outweigh the costs.
When data is collected from a large sample of people by using questionnaires this is called a major sex survey.
The first major sex survey was done by Alfred C. Kinsey between the late 1930s and 1940s. Even though the results are of historical interest, the research is a good example. The sample of the research was obtained by interviewing 5,300 males. Their responses were reported in a scale, called the sexual behaviour in the human male (1948). Females contributed to the sexual behaviour in the human female (1953). They chose not to use probability sampling because of non-response problems. However, there is currently no information about the accuracy of the used sample. College students, well-educated people, young people, protestants, people from Indiana and people from the Northeast were over presented. He used face-to-face interviewing by Kinsey himself or his well-trained colleagues. They cross-checked the participant’s reports for false information. The sample contained a unlikely high level of homosexuals and people with unusual sexual practices. Therefore, it is impossible to say how accurate the Kinsey statistics are.
The National Health and Social Life Survey (NHSLS) used slipshod sampling. A large sample of the national population was needed to say what sexual behaviour looked like in America. The researchers used probability sampling for this purpose. It excluded only three percent of the population, which were institutionalised people. Data was obtained with interviews in addition to a written questionnaire. The NHSLS is one of the best surveys because outstanding sampling -and interview techniques were used.
The National Survey of Sexual Health and Behaviour (NSSHB) is the most recent major survey done in the United States. A probability sample was used and 5,865 people participated. The method they used for sampling, which was based on the random selection of phone numbers was excellent. However, it caused the response rate to be low. Therefore, it is not possible to be sure the result can be generalised to the whole population. They obtained a good sample size and ethnic diversity though.
Every ten years, a major sex survey is conducted in Britain called the National Survey of Sexual Attitudes and Lifestyle (Natsal). They use postal codes for sampling, which is an excellent method and the sample size is good (about 15,162 completed interviews). Interviewers went to participants homes to complete the survey. Also, urine samples were collected to test for sexual transmitted diseases. Another major research was done in Australia where computer-assisted telephone interviews were used to study health and relationships. Their findings were that the age of first intercourse has decreased over the last couple of years which is consistent with findings in the United States.
Latinas in the United States were interviewed to assess patterns of interpersonal victimization. This included sexual assault and violence by one’s partner. The study was called SALAS, meaning Sexual Assault among Latinas. Interviews were conducted over the phone in either English or Spanish. It used many methods suitable for investigating among minority populations. Sex research in minorities indicates sensitive issues and need additional consideration in language and ethnicity of the interviewer.
Magazine surveys are using samples that our out of their control. They are distributed among different clienteles and only to readers of the magazine. The response rate is about three percent, which is not even generalizable to the population of people reading the magazine. Ethnic backgrounds, marital status and age are unknown. These details are crucial in evaluation of the outcomes. Therefore, it would not be legitimate to infer conclusions based upon these statistics.
An example of a study of a special population is the Coxon Study of gay men in the AIDS era. The project was called SIGMA and was used to understand the behaviour of men in the AIDS era. Participants were divided into two groups, one was the “out” and easy to administer gay people that were recruited in bars and pubs. The second, more difficult group consisted of “hidden” gays. After that respondents were asked to nominate persons in the other category. This technique of sampling is called snowball sampling or respondent-driven sampling. The study used the daily diary method where participants reported their sexual experiences. It resulted in a richer and contextualized overview of people’s sexual experiences. However, memory in self-reports is a major concern.
A whole new era in sex research began by administering web-based surveys. Advantages were that larger and broader samples could be recruited. They also have an advantage in studying specific populations defined by their sexual behaviour. Especially when studying taboo behaviour. They also have the advantage to eliminate extraneous factors, such as the ethnicity of the interviewer. A disadvantage is that the researcher does not have any control over the environment. Also, individuals might respond multiple times or sabotage the interview with fake answers.
The influence of mass media on sexuality can be analysed by content analysis, it consists of a set of procedures that are utilised to make valid inferences about text. It can be any text, an article in a magazine, a rap song or a romance novel. Again, sampling becomes an issue. The population must be defined an whether you want to analyse over a specific time period. Then, a coding protocol should be established. One important consideration is intercoder reliability, the correlation or percentage of agreement between different coders rating the same texts. It is a powerful scientific technique that allows people to know how the media is portraying sexuality.
Most research discussed so far uses quantitative measures, however qualitative measures can give useful insights into the subject as well. Qualitative research is done in a naturalistic, holistic manner which includes observation and in-depth interviewing of a participant. The results are conveyed into words instead of numbers. It seeks a complete picture of the participant and the context. Ethnography is a qualitative method that is used to provide a description of a social setting, society or human group. Participant-observer techniques include methods where the researcher becomes part of a community in order to observe its behaviour.
The research discussed so far looks at natural occurrence of sexual behaviour. Such research is called correlational, meaning that the researcher does not manipulate variables, but studies the natural occurrence of relationships between variables. An experiment is another type of research where the experimenter studies an independent variable which is manipulated and its dependent variables that occur as a result of the manipulation of the independent variable. All other factors are held constant. The researchers is then permitted to make causal inferences about the behaviour observed in his participants.
Meta-analysis is a technique to create order among different studies on the same topic. It is a statistical technique where results from different studies are combined to determine what, taken together these studies say. There are three steps that need to be taken:
The researcher must locate all the previous studies based on the question that needs to be investigated.
For each study, the researcher must compute a statistic to measure the direction and difference between males and females. This is the value of d.
The researcher must average all the values of d over all the studies.
There are several important statistical concepts to understand before understanding reports of sex research.
Average is used to summarise data. The mean is the average of scores of the respondents. The median is the middle score.
Variability is an indication of the average of all the respondents. This can be used to judge the numbers in a single respondent or group of respondents.
Average versus normal is an important consideration, conclusions based on a single average number are often mistaken. It is important to note that variability in (sexual) behaviour should be considered in classifying normal behaviour.
Incidence versus frequency is another important consideration. Incidence refers to the percentage who engaged in a specific behaviour. Frequency is how often they do this. Cumulative frequency is the percentage of people who engaged in specific behaviour before a certain age.
Correlation is the number that can be used to measure a relationship between two variables.
Summary with all chapters of the 13th edition of Understanding Human Sexuality by Hyde & Delamate
Summary with the mandatory readings for the course Sexology (a free elective for Psychology students at the UvA).
Selected by VSPA
JoHo can really use your help! Check out the various student jobs here that match your studies, improve your competencies, strengthen your CV and contribute to a more tolerant world
There are several ways to navigate the large amount of summaries, study notes en practice exams on JoHo WorldSupporter.
Do you want to share your summaries with JoHo WorldSupporter and its visitors?
Field of study
Add new contribution