Research Methods

Likert scales are used in questionnaires where participants indicate their level of agreement with a statement, such as strongly agree, agree, neutral, disagree, or strongly disagree.

Rating scales allow participants to indicate the strength of their feelings, for example from very funny to not funny at all.

Fixed choice questions provide participants with a set of possible answers from which they choose.

Open questions allow participants to answer freely in their own words, producing qualitative data. They provide rich, in-depth responses and allow a deeper understanding of individual views. However, they are more difficult and time-consuming to analyse and take longer to administer.

Closed questions provide fixed answers such as yes/no or scale responses, producing quantitative data. They are easier and quicker to analyse but provide less detail and are more prone to response bias.

Questionnaires can be distributed to large samples easily and efficiently. However, participants may not answer truthfully due to social desirability bias, and response bias may occur if participants consistently answer in a particular way.

Structured interviews use pre-determined questions asked in a fixed order. They are easy to replicate and analyse, reduce interviewer bias, and increase inter-interviewer reliability. However, they restrict responses and reduce depth of data because the interviewer cannot explore responses further.

Unstructured interviews do not use pre-determined questions. Instead, they are guided by a general aim, and participants are encouraged to speak freely. This provides detailed and in-depth responses. However, they are more susceptible to interviewer bias, require well-trained interviewers, are more expensive, and are time-consuming to analyse.

An interview schedule is the set of questions used in interviews. This should be standardised to reduce bias.

If language is too complex or contains jargon, participants may not understand questions, reducing validity. Leading questions may influence responses. Interviewer bias may occur through tone of voice, facial expressions, or body language.

Independent and Dependent Variable

The independent variable is the variable that is manipulated by the researcher in order to observe the effect it has on another variable. The dependent variable is the variable that is measured as the outcome of the manipulation of the independent variable. The dependent variable should change as a result of the manipulation of the independent variable.

Operationalisation

Operationalisation refers to the process of defining variables so that they can be measured in a clear and precise way. This ensures that the research is objective and that the study can be replicated by other researchers. For example, if the dependent variable is aggression, it may be operationalised as the number of punches thrown during a set period of time.

Extraneous Variables

An extraneous variable is any variable other than the independent variable that may affect the dependent variable if it is not controlled. These variables can reduce the validity of the study because they make it unclear whether the independent variable is responsible for the results.

Participant variables are differences between individuals that may influence the outcome of a study. These include factors such as age, gender, mood, intelligence, personality, memory, and prior experience.

Situational variables are differences in the environment in which the study takes place. These include factors such as temperature, noise levels, time of day, and order effects.

Investigator variables are differences caused by the behaviour of the researcher. These include factors such as tone of voice, body language, or leading questions that may influence participants’ responses.

Participant variables can be controlled by using a large and representative sample, randomly allocating participants to conditions, or using repeated measures or matched pairs designs.

Situational variables can be controlled through standardised procedures, ensuring that all participants experience the same conditions, and through counterbalancing to reduce order effects.

Investigator variables can be controlled using single-blind or double-blind procedures, as well as randomising conditions and ensuring the researcher does not influence participants.

Demand characteristics are cues within a study that may allow participants to guess the aim of the research. As a result, participants may change their behaviour, which reduces the validity of the findings.

Investigator effects refer to any unintended influence of the researcher on the outcome of the study. For example, the researcher’s tone of voice, facial expressions, or body language may influence how participants behave or respond.

Random allocation involves randomly assigning participants to different conditions in order to distribute participant variables evenly across groups.

Counterbalancing is used in repeated measures designs to control order effects. Half of the participants complete the conditions in one order, while the other half complete them in the reverse order.

Randomisation involves presenting materials or conditions in a random order to reduce bias.

Standardisation involves keeping all aspects of the procedure consistent for all participants to improve reliability.

Control groups are used as a comparison group that does not receive the treatment or independent variable. This allows researchers to determine the effect of the independent variable.

British Psychological Society’s Code of Ethics

The British Psychological Society outlines four key ethical principles. Respect involves valuing the dignity and rights of participants. Competence requires researchers to have the appropriate training and skills. Responsibility involves ensuring that participants are not harmed and that the researcher acts professionally. Integrity requires honesty and accuracy in conducting and reporting research.

Ethical Issues

Informed consent means that participants must be given enough information to make an informed decision about whether to take part. For participants under the age of 16, consent must be obtained from a parent or guardian. If it is not possible to obtain consent, presumptive consent or prior general consent may be used.

Deception involves misleading participants about the aims or procedures of the study. It should only be used when necessary and must be followed by a debrief to explain the true purpose of the research.

The right to withdraw means that participants can leave the study at any time and can withdraw their data after participation. This must be made clear during the consent process and debriefing.

Protection from harm requires that participants should not be exposed to physical or psychological risk greater than they would experience in everyday life.

Confidentiality means that participants’ data should not be identifiable. Researchers should use codes or numbers instead of names when reporting findings.

Privacy refers to a participant’s right to control information about themselves. Observations should only take place where individuals would reasonably expect to be observed.

What is Peer Review?

Peer review is the process in which other experts in the same field evaluate research before it is published. This ensures that the research is of high quality, accurate, and credible. The reviewers are usually anonymous and objectively assess the study.

Aims of Peer Review

Peer review is used to help allocate research funding, as experts decide which studies are worthy of financial support.

It also validates the quality and relevance of research by assessing the methodology, statistical analysis, and conclusions of a study.

Peer review allows reviewers to suggest improvements or revisions, or in some cases reject research that is of poor quality.

It is also used to assess the research quality of academic institutions, such as through the Research Excellence Framework.

Evaluation of Peer Review

One issue is that it can be difficult to find a suitable expert for the review process.

Although reviewers are anonymous, this may lead to bias, as some may criticise competing researchers unfairly.

Publication bias may occur, as journals may favour research with positive or significant results rather than studies with negative findings.

Peer review may also preserve the status quo, as research that challenges established theories may be less likely to be published.

Psychological research can have significant effects on the economy.

Research into mental health can reduce workplace absenteeism, as effective treatments improve wellbeing and productivity.

Research into family roles, such as the importance of fathers, has influenced policies such as parental leave and flexible working arrangements, allowing both parents to remain economically active.

Research into criminal behaviour can reduce costs associated with crime by improving policing methods and developing effective interventions for offenders.

Reliability refers to the consistency of a measure. A measure is reliable if it produces the same results when repeated.

Internal reliability refers to consistency within a test, meaning that items within the test measure the same thing. External reliability refers to consistency over time.

Test-retest reliability involves administering the same test twice and comparing the results. A high positive correlation indicates good reliability.

Inter-observer reliability involves comparing the observations of different researchers. A strong agreement indicates high reliability.

Reliability can be improved by removing or rewriting ambiguous questionnaire items, using structured interviews, training observers, and standardising procedures.

Validity refers to the extent to which a study measures what it intends to measure.

Internal validity refers to whether the observed effects are due to the independent variable rather than extraneous variables.

External validity refers to whether the findings can be generalised to other situations, people, or time periods. This includes ecological validity and temporal validity.

Face validity refers to whether a measure appears to assess what it is supposed to measure.

Concurrent validity involves comparing a new test with an established test to see if similar results are obtained.

Validity can be improved by using control groups, standardised procedures, blind designs, and triangulation of methods.

Scientific research must be objective, meaning it is based on facts rather than personal opinions. It must use empirical methods, meaning that evidence is obtained through observation or experimentation.

Research must be replicable so that findings can be verified by other researchers.

Theories are developed through observation and tested using hypotheses. This process involves drawing conclusions, proposing theories, and conducting further studies.

Falsifiability means that a theory must be capable of being disproven. According to Popper, scientific theories should be tested in a way that attempts to disprove them rather than confirm them.

Replication is essential for scientific credibility, as repeated findings increase confidence in results.

Paradigms are shared sets of assumptions within a scientific discipline. A paradigm shift occurs when new evidence challenges existing beliefs and leads to a change in understanding.

Pilot studies are small-scale studies conducted before the main research to identify and correct potential issues.

Psychological reports are written to communicate research findings.

The abstract provides a brief summary of the entire study, including aims, methods, results, and conclusions.

The introduction reviews existing research and outlines the aims and hypotheses.

The method section describes how the study was conducted, including design, participants, materials, procedure, and ethical considerations.

The results section presents the findings using descriptive statistics and, where appropriate, inferential statistics.

The discussion interprets the findings, evaluates the study, and suggests improvements and applications.

Referencing involves listing all sources used to avoid plagiarism and allow others to locate the original material.

Descriptive Statistics

Measures of central tendency include the mean, median, and mode. The mean is calculated by adding all values and dividing by the number of values. The median is the middle value in an ordered set, and the mode is the most frequent value.

Measures of dispersion include the range and standard deviation. The range is the difference between the highest and lowest values, while standard deviation measures how spread out data is around the mean.

Mathematical Skills

Psychologists must be able to calculate and interpret fractions, decimals, percentages, ratios, and standard form.

Percentages are calculated by dividing a value by the total and multiplying by 100.

Standard form expresses numbers using powers of ten.

Significant figures involve rounding numbers to a specific level of precision.

Presentation and Display of Quantitative Data

Data can be presented using raw data tables and frequency tables.

Bar charts are used for categorical data and have separate bars.

Histograms are used for continuous data, and bars touch because the data is on a continuous scale.

Line graphs and frequency polygons show changes over time or comparisons between conditions.

Scattergrams are used to display correlations between variables.

Pie charts represent proportions of a whole.

Distributions include normal distributions, where data is evenly spread, and skewed distributions, where data is unevenly distributed.

Probability and Significance

Probability refers to the likelihood that results occurred by chance. In psychology, a probability level of 0.05 is typically used, meaning there is a 5 percent chance that results occurred randomly.

Significance refers to whether results are unlikely to have occurred by chance.

Statistical tables are used to compare calculated values with critical values to determine significance. This requires knowledge of whether the hypothesis is one-tailed or two-tailed, the number of participants, and the significance level.

Type I errors occur when the null hypothesis is incorrectly rejected. Type II errors occur when the null hypothesis is incorrectly accepted.

Statistical Testing

The sign test is used when analysing differences in related designs with nominal data.

The researcher calculates the number of positive and negative differences between conditions and identifies the less frequent sign as the calculated value.

This value is then compared to a critical value to determine significance.

A statement of significance must include whether the result is significant, the observed value, the critical value, the number of participants, and whether the hypothesis is accepted or rejected.

Choosing a Statistical Test

The choice of statistical test depends on whether the study is testing a difference or a relationship, whether the design is related or unrelated, and the level of measurement.

Nominal data uses tests such as the sign test or chi-squared.

Ordinal data uses tests such as Mann-Whitney or Wilcoxon, and correlations use Spearman’s rho.

Interval data uses parametric tests such as t-tests and Pearson’s r.