Revise Research methodology for Cambridge 9990 (3.1.2)

The experimental method is the cornerstone of psychological research, designed to establish cause-and-effect relationships. It involves the deliberate manipulation of an Independent Variable (IV) to observe its effect on a Dependent Variable (DV), while rigorously controlling all other extraneous variables. For example, a researcher might manipulate the type of music played (IV) to measure its effect on task performance (DV). There are three main types: laboratory, field, and natural/quasi-experiments. Laboratory experiments offer the highest level of control, enhancing internal validity, but may suffer from low ecological validity due to their artificial setting. Field experiments balance control with realism by taking place in a natural environment, while natural experiments study the effects of a naturally occurring IV that the researcher cannot manipulate.

Observations involve watching and recording behaviour. They can be structured, using pre-determined behavioural categories and a coding scheme, or unstructured, where the researcher records all relevant behaviour without a system. The observer's role is also key. In a participant observation, the researcher becomes part of the group being studied, whereas in a non-participant observation, they remain separate. Furthermore, observations can be overt (participants know they are being watched) or covert (participants are unaware). Covert, participant observations can yield highly valid data as they reduce demand characteristics, but they raise significant ethical issues regarding deception and informed consent. Structured observations produce quantitative data that is easy to analyse, while unstructured observations provide rich, qualitative detail.

Self-report methods gather data by asking participants about their thoughts, feelings, or behaviours. Questionnaires use written questions and can include closed questions (e.g., yes/no, rating scales like Likert scales) which generate quantitative data, or open questions which generate detailed qualitative data. Interviews are face-to-face or remote conversations. Structured interviews use pre-set questions, making them standardised and replicable. Semi-structured interviews have some prepared questions but allow the interviewer flexibility to ask follow-up questions to explore responses in more depth. A major weakness of all self-reports is the potential for social desirability bias, where participants give answers they believe are more socially acceptable, reducing the validity of the data.

Once data is collected, it must be analysed and presented. Quantitative (numerical) data is summarised using descriptive statistics. Measures of central tendency (mean, median, mode) describe the 'typical' score, while measures of dispersion (range, standard deviation) describe the spread of scores. The mean is the arithmetic average, the median is the middle value, and the mode is the most frequent value. The standard deviation is a powerful measure of how scores cluster around the mean. Qualitative (non-numerical) data, such as interview transcripts, is often analysed using thematic analysis to identify recurring patterns and themes. Data can be presented visually in graphs, such as bar charts for discrete categories or histograms for continuous data, to make patterns easier to understand.