Reliability in psychological research

Reliability refers to the consistency and replicability of a study's findings. A reliable study produces similar results when repeated under the same conditions, often due to standardised procedures that control extraneous variables ~ factors that could influence the results but are not the focus of the study.

Factors that contribute to reliability

• Standardised procedures - Using the same methods, materials, and timings for all participants ensures the study can be replicated accurately.
• Control of extraneous variables - Keeping conditions consistent, such as location or environmental factors, reduces variability in results.
• Consistency with other studies - If similar research yields comparable findings, this supports the reliability of the original study.

Evaluating reliability in Loftus and Palmer (1974)

• Aim - Loftus and Palmer (1974) investigated how leading questions affect eyewitness testimony (EWT) - the recall of events by witnesses, which can be influenced by wording in questions.
• Method - This was a laboratory experiment where participants watched standardised videos of car accidents and answered questions with varying verbs (e.g., "smashed" versus "bumped").
• Strengths for reliability - The procedure was highly standardised, with controlled timings, identical videos, and a consistent lecture hall setting, making it easy to replicate. Extraneous variables like weather were eliminated. Similar studies have confirmed that verbs influence recall, showing consistent findings.

Evaluating reliability in Yuille and Cutshall (1986)

• Aim - Yuille and Cutshall (1986) examined EWT in a real-life gun shop robbery, interviewing witnesses months later to compare with initial police statements.
• Method - This was a field study using real witnesses from a naturally occurring event, with interviews replicating police procedures.
• Strengths for reliability - Coding units for analysing interviews were objective and rigorous, leading to consistent results when compared to police transcripts. The replication of police interview details added some reliability.
• Limitations for reliability - The unique, traumatic event could not be repeated exactly, lacking standardisation. Witnesses had different vantage points, introducing participant variables ~ individual differences that might bias results. Only 13 witnesses were studied, and potential media influence over months could have altered testimonies, reducing replicability.

Validity in psychological research

Validity measures how well a study accurately reflects or assesses what it claims to investigate. High validity means the findings truly represent real-world phenomena, while low validity might stem from artificial settings or uncontrolled variables.

Ecological validity

Ecological validity refers to the extent to which findings can be generalised to real-life settings, considering natural behaviours and environments.

Evaluating validity in Loftus and Palmer (1974)

• The artificial setting lacks ecological validity, as watching staged videos does not match the emotional intensity or spontaneity of real accidents.
• Participants might have guessed the study's aims, leading to demand characteristics - changes in behaviour due to awareness of being studied.
• Questionnaires do not reflect the real consequences faced by actual witnesses, limiting applicability to natural settings.

Evaluating validity in Yuille and Cutshall (1986)

• As a real-life event with genuine witnesses, it has high ecological validity, capturing natural behaviours and uncontrolled variables typical of actual crimes.
• Findings more accurately represent spontaneous EWT in traumatic situations.

Objectivity in psychological research

Objectivity means conducting research without personal bias, using scientific methods to establish cause-and-effect relationships. This involves careful controls, replicable procedures, and quantitative data ~ numerical information that can be statistically analysed for credibility.

Achieving objectivity in different methods

• Laboratory and field experiments - These maintain objectivity through strict controls over variables and standardised procedures, minimising experimenter effects ~ unintended influences from the researcher.
• Case studies - These can introduce subjectivity, especially when working closely with participants, potentially compromising impartiality.

Evaluating objectivity in Loftus and Palmer (1974)

• Strengths for objectivity - The study used scientific principles with controls like consistent film timing and length, producing quantitative data on speed estimates and recall. This setup minimises bias and supports objective, replicable conclusions about leading questions in EWT.

Evaluating objectivity in Yuille and Cutshall (1986)

• Strengths for objectivity - The replication of police interviews and use of objective coding for data analysis added scientific rigour, aligning with principles of cause-and-effect investigation.
• Limitations for objectivity - Working closely with a small group of witnesses risks subjectivity, such as researchers' interpretations influencing findings, even if unintentional.

Credibility in psychological research

Credibility refers to the believability and trustworthiness of research findings, often based on whether the information comes from reputable sources and follows scientific standards. High credibility requires objectivity, replicability, and validity, along with the researcher's expertise.

Factors influencing credibility

• Source quality - Findings from peer-reviewed journals are more credible than unverified online sources.
• Methodological rigour - Studies that are objective, replicable, and valid enhance trustworthiness.
• Researcher expertise - Experts like Elizabeth Loftus, known for EWT research, add credibility through their established reputation.

Comparing credibility across methods

• Laboratory and field experiments - These generally have higher credibility due to adherence to scientific protocols, producing reliable and valid data.
• Case studies - These may have lower credibility as they involve subjective elements and are harder to replicate.

Evaluating credibility in key studies

Loftus and Palmer (1974) has high credibility stemming from Loftus's expertise, objective procedures, and replicable design yielding valid quantitative data. Published in reputable journals, it establishes believable insights into how leading questions distort EWT.

Ethics in psychological research

Ethics involve moral principles guiding research to protect participants, including obtaining consent, avoiding deception where possible, and ensuring no harm. Studies must balance scientific value with participant welfare, following guidelines like those from the British Psychological Society.

Key ethical principles

• Informed consent - Participants should agree to take part after understanding the study's nature.
• Protection from harm - Researchers must minimise physical or psychological distress.
• Deception - If used, it should be justified and followed by debriefing ~ explaining the true purpose post-study.
• Right to withdraw - Participants can leave at any time.

Evaluating ethics in Loftus and Palmer (1974)

• Strengths for ethics - Participants gave consent to view videos and answer questions, knowing key elements of the procedure.
• Limitations for ethics - Deception occurred through misleading verbs (e.g., "smashed" implying greater impact) without participants' awareness, potentially causing distress from viewing car crashes. No mention of debriefing to clarify the deception.

Evaluating ethics in Yuille and Cutshall (1986)

• Strengths for ethics - Informed consent was obtained, with some participants opting out due to stress, showing respect for their well-being. As a naturally occurring event, no artificial distress was induced, providing better protection from harm.
• Limitations for ethics - Deception was present in misleading questions (e.g., about non-existent details like a "yellow/blue quarter panel"), without full awareness. Reliving a traumatic event months later could cause psychological harm, though the study allowed withdrawals.