Psychology Unit 4 - Research Methods

The Major Features Of Science

• Objectivity: Scientists stove to be objective in their observations and measurements. Should not let personal prejudices/emotions influence their theory, explanations, findings etc.
• Replicability: It must be able to be repeated. All details of the original study must be published. Replicability tends to be greatest when experiments are conducted in a controlled way.
• Control: Demonstrate casual relationships to enable to predict our world. Vary one factor and observe its effect on the dependent variable. Everything else must be kept the same.
• Empirical Methods: Collecting data through direct observations/experiment. Opposite is rational research, where a theory is constructed from reasoned argument rather than direct observation.
• Theory Construction: Aim is to record facts and to use facts to construct theories to help us understand and predict the natural phenomena around us. A theory is a collection of general principles that explain observations and facts.
• Hypothesis Testing: Theories/models are tested by generating specific, testable hypotheses from the theory. To investigate whether evidence supports the theory or the theory needs to be modified in the light of the findings.

Conventions For Reporting Investigations

• Abstract: A summary covering all the main headings-normally about 150 words. They are important because the reader can see all of the study at a glance.
• Introduction: A review of previous research leading to a rationale for why the researchers intend to conduct this study. At the end, the aims/hypothesis are stated.
• Method: Design=Justification of methods+ethics. Participants=Population+sampling method. Materials=What was used. Procedure=How the study was carried out, replicability.
• Results: What the researcher found=descriptive statistics (graphs/tables/averages), inferential statistics and/or qualitative data.
• Discussion: Researchers offer explanations of the behaviours they observed and might also consider the implications of the results and suggestions for future research.
• References: The full details of any journal articles or books they have mentioned.

New Knowledge & Peer Review

• Process: Subjecting a piece or research to independent scrutiny by other psychologists working in a similar field who consider the research in terms of its validity, significance and originality. Researcher sends their report to the editor of the journal, who sends it to 3 referees-analyse/comment on whether they should include it in the publication.
• Purpose: System of quality assurance applies to research publication-to ensure its of high quality. Peer review provides a way of checking the validity, making a judgement about the credibility and assessing the quality and appropriateness of the design and methodology.
• Pitfalls: Reports in journals tend to be about findings that support hypotheses=file drawer problem. Also, issue of bias . Theres a tendency for certain institutions to get published more. Referees are not paid, it looks good on their CV.
• Why is it important to get published in a peer review journal?: Articles that are published in peer reviewed journals are of greater merit. They will only get published if it makes an important contribution to the scientific field, and the research process has been methodologically and ethically sound.

Different Research Methods

• Lab Experiment: Controlled setting-IV is manipulated+DV is measured. Researcher has high control over extraneous variables+cause and effect can be established.
• Field Experiment: Natural environment-IV is manipulated+DV is measured. Extraneous variables are hard to control, cause and effect can be established.
• Natural Experiment: Naturally occurring-IV is naturally occurring+DV is measured. No random allocation, extraneous variables are not controlled.
• Case Study: In depth area of study-one person/small group-involves several other methods and looks at the personal experiences of the participants.
• Observation: Researcher does not get involved, have behavioural categories and will watch and record-whole event or intervals. If they are aware=overt, unaware=covert
• Interview: Gather detailed information-small groups-structured, non-structured.
• Questionnaires: Report on their own behaviour(face to face/online/post)-open or closed-for a particular group.
• Correlation: Relationship between 2 things-positive/negative/zero.
• Content Analysis: Counting technique, how many times something occurs. Used to analyse transcripts, researchers may have a coding system.

Sampling Methods

• Opportunity Sampling: Participants are selected by using those people who are most easily available. Bias=High, due to availability. Generalisability=Low, unlikely to represent the target population. It is quick, easy and cheap.
• Volunteer Sampling: The researcher will advertise and then the participants will contact the researcher. Bias=High, only attracts certain people. Generalisability=Low, unlikely to represent the target population. Participants want to take part.
• Random Sampling: All members of the target population are identified, participants are then randomly selected. Bias=Low, everyone has an equal chance of being selected. Generalisability=High, may not be representative. It is time consuming and expensive.

Reliability: Consistency

• Internal: Measure of the extent to which something is consistent within itself.-Inside.
• 1) Split Half Method; Compare a persons performance on 2 halves of a questionnaire/test.
• External: Measure of consistency over several different occasions.-Outside.
• 1) Test-Retest Method; Person is given a questionnaire/test/interview on one occasion and is repeated after a reasonable interval.
• 2) Observer Reliability Method; Extent to which different observers achieve similar results when observing the same participants.
• Improving Reliability:
• 1) Lab; Make sure variables are operationalised and tightly control for extraneous variables. Materials should be clearly described. Method should be in depth and be included in the report so it can be replicated.
• 2) Interview/Questionnaire; Pilot study should be carried putto identify any problems. Questions should be checked to ensure they are clear and non ambiguous. When carrying out an interview, researchers should be trained.
• 3) Observation; Use more than one observer and compare observer reliability. Observers should be trained and know how to use the coding system. Behavioural checklist should be clear and non ambiguous. Data can be checked by sliming the observation.

Validity

• Internal: Researcher tested what they intended to.-Inside.
• 1) Face Validity; Whether at face value the study appears to measure what it set out to.
• 2) Concurrent Validity; Compares a new test to other tests to see if it produces similar results.
• 3) Predictive Validity; Measure is valid because it can accurately predict the performance of that behaviour.
• External: Extent to which the researchers findings can be generalised to other situations.-Outside.
• 1) Population Validity; Results can be generalised to other groups of people.
• 2) Ecological Validity; Findings can be generalised to situations outside the research setting.
• Improving Reliability:
• Internal- 1) Lab; Extraneous variables need to be controlled. Variables must be operationalised. IV/DV may need to change=more accurate.
• 2) Field; Extraneous variables need to be controlled. Variables must be operationalised. IV/DV may need to change=more accurate.
• 3) Observations; More than one observer must be used and should be trained. Checklist may need improving.
• External- 1) Lab; Sampling technique may need changing. IV/DV must be true to life. Increase sample size.
• 2) Field; Sampling technique may need changing, increase sample size. Participants must not know it is a study, it will affect their behaviour.
• 3) Observation; Observe as many people as possible. Use covert observations.

Ethics

• Informed Consent: Researchers should always gain the valid consent.-By participants signing a consent form. It should include comprehensive information concerning the nature and purpose of their research and their role in it.
• Protection From Harm: Researchers have primary responsibility to protect participants from negative physical and mental harm.-Study should be stopped if any risks greater than those that would be expected to be experienced in real life are suspected.
• Deception: Withholding of information or the misleading of participants is unacceptable.-Must be fully debriefed after the study.
• Confidentiality: Privacy of participants and any information gained is kept confidential. Identities should never be revealed.-After research is published, information should be locked away or destroyed.
• Right To Withdraw: Participants must be aware that they can withdraw at anytime without giving a reason. They can decline to answer questions or take part in certain aspects of a study.-They should be informed at the start of the study of this right.

Graphs, Designs & Measurements

• 1) Bar Chart: Nominal.-A set of vertical bars with a space between them. Bars represent different categories.
• 2) Histogram: Ordinal/Interval/Ratio.-No space between each bar. Bars are continuous. Units of measurement are on the x axis.
• 3) Frequency Polygon: Ordinal/Interval/Ratio.-Alternative to a histogram. Useful when you need to show 2 sets of data on the same graph.
• 4) Scattergram: Ordinal/Interval/Ratio.-Shows the relationship between 2 variables. Continuous data. Checks reliability.
• 1) Independent Measures: Group of participants do condition 1 and anther group does does condition 2.
• 2) Repeated Measures: Same group of participants do each condition.
• 3) Matched Pairs: Participants in group 1 and 2 are different but have similar characteristics.
• 1) Nominal: Data falls into separate categories. Each piece of data can only belong in one category. Simply count the number of participants that belong to each category.
• 2) Ordinal: The data is ordered in some way, e.g. attitude scales.
• 3) Interval/Ratio: Data is measured using units of equal intervals. Each unit has an agreed value.

Probability & Significance

Slide 11