PSYA1 – Research Methods



PY2 and PY3 – Research Methods

Use the table below to define key terms, and other two mark answers that you will need for research methods questions.

|Method |Definition |

|Laboratory Experiment |An experiment carried out in a controlled setting, which allows the experimenter to control |

| |extraneous variables and so have more confidence that the IV is causing any change in the DV|

|Field Experiment |An experiment carried out in a more natural setting outside the lab. The experimenter still |

| |manipulates the IV but has less control over extraneous variables, so can’t be so sure about|

| |causality. On the other hand, participant behaviour is more likely to be natural. |

|Quasi Experiment |An “almost” experiment that lacks one of the features of a true experiment – such as a |

| |difference study or a natural experiment. |

| |An example of a natural experiment would be comparing rates of mental illness before and |

| |after an earthquake. |

|Natural Experiment |An experiment where the IV occurs naturally and the experimenter measures a change in the |

| |DV. The experimenter has no control over extraneous variables. An example of a natural |

| |experiment would be comparing rates of mental illness before and after an earthquake. |

|Correlation Study |Research that studies the strength of connection between two variables. If the both increase|

| |together, that is a positive correlation. When on increases and the other decreases this is |

| |a negative correlation. We cannot assume that one variable is causing the other – eg ice |

| |cream sales and shark attacks are positively correlated. |

|Difference studies |A quasi experiment where the IV is not really altered (it just exists eg gender) Sheridan |

| |and King measured male and female obedience when asked to shock puppies.The Iv is gender |

| |which occurred naturally |

|Naturalistic Observation |A research method that takes place in a natural environment where behaviour is simply |

| |observed and no variables are controlled. Eg Diane Fossey observing mountain gorillas in the|

| |wild. |

|Controlled Observation |Research that takes place in an environment where certain variables are controlled by the |

| |observer. Eg Milgram’s research on obedience involved controlling certain variables, such as|

| |the prompts given when participants wanted to stop. |

|Self Reported Data |Any method where participants are asked to report their own attitudes, abilities or feelings|

| |such as a questionnaire or interview. Eg Buss asked participants about what they would look|

| |for in a life partner. |

|Questionnaire |A written list of questions completed by a participant on their own. The question may be |

| |open questions “What do you think about smoking?” or closed “Do you smoke – yes or no” |

|Interview |A face-to-face interaction between researcher and participant. The interview can be |

| |structured (like reading out a questionnaire) semi-structured (where questions can be |

| |followed up depending on the participant’s response) or unstructured, where the researcher |

| |states the aims and a discussion takes place. |

|Content Analysis |A kind of indirect observational study where the researcher draws conclusions about |

| |behaviour based on written or verbal material. For example, we could study attitudes to |

| |sexism based on the content of TV adverts. |

|Case Study |A research investigation that involves a detailed study of an individual, group, institution|

| |(eg a school) or an event. Case studies produce rich and detailed information about |

| |participants, but are hard to generalise from. |

|Method |Advantages and Disadvantages |

|Laboratory Experiment |Advantage: Control over extraneous variables mean we can be confident that any change in |

| |the DV is caused by the IV. |

| |Disadvantage: High levels of control make the situation artificial, which means participants|

| |may behave differently from real life, causing ecological validity to be low |

|Field Experiment |Advantage: Participants are in their natural environment which means that their behaviour |

| |will be less influenced by demand characteristics and so will be more generalizable. |

| |Disadvantage: The experimenter has less control over extraneous variable, so wil be less |

| |sure that the IV caused the change in the DV. |

|Natural Experiment |Advantage: It allows us to research topics where the IV cannot be manipulated for ethical or|

| |practical reasons eg Earthquakes |

| |Disadvantage: Because the IV is not manipulated we cannot demonstrate causal links. Also, |

| |extraneous variables are not controlled either, making it difficult to draw causal |

| |conclusions. |

|Correlation Study |Advantage: If a correlation is found between variables then it is worth investigating a |

| |causal relationship by doing an experiment. |

| |Disadvantage: A correlation between two variables does not suggest that one cause the other |

| |a third variable may be responsible eg increased ice cream sales don’t cause shark attacks. |

|Naturalistic Observation |Advantage: Naturalistic observation gives us a picture of how people behave in ordinary |

| |settings, so produces high ecological validity. Eg – littering behaviour: it is more |

| |accurate to watch littering behaviour than give out questionnaires. |

| |Disadvantage: There is little control of extraneous variables which means that the observer |

| |cannot be sure about what is causing the behaviour. |

|Controlled Observation |Advantage: Control over extraneous variables mean we can be more confident about what is |

| |causing the behaviour than we would with a naturalistic observation. |

| |Disadvantage: High levels of control make the situation artificial, which means participants|

| |may behave differently from real life, causing ecological validity to be low |

|Self Reported Data |Advantage: The participant is giving their own explanation of the behaviour involved, or |

| |telling us their attitude about something. |

| |Disadvantage: Participants may not tell us the truth for a range of reasons – to impress |

| |(social desirability bias), because they can’t remember or simply because they don’t know. |

|Questionnaire |Advantage: Compared to an interview, we can work much faster – lots of participants can |

| |complete a questionnaire at the same time, so we can collect much more data in the same |

| |amount of time |

| |Disadvantage: The method is limited to people who can read and who have the time and |

| |motivation to complete the form. |

|Structured Interview |Advantage: Easily replicated because it follows standardised procedures – the interviewer |

| |ask the same questions in the same order with no follow-ups. |

| |Disadvantage: Reliability could be low though, if the researcher behaves differently for |

| |each participant, or if more than one researcher is used. |

|Semi-Structured Interview |Advantage: A skilful researcher can ask follow-up questions that can provide rich detailed |

| |information about each participant. |

| |Disadvantage: The expectations of the interviewer may influence their interpretation of the |

| |participant’s answers – this is known as interviewer bias. |

|Content Analysis |Advantage: High ecological validity, because it is based on what people actually do – they |

| |are not aware that their work is being studied so their won’t be demand characteristics. |

| |Disadvantage: Observer bias – each observer may be influenced by their expectations and only|

| |see what they are looking for. Eg they may notice sexism in car adverts but not in furniture|

| |ads. |

|Case Study |Advantage: You can use a variety of methods to get rich detailed information about the |

| |individual/group or institution you are studying. |

| |Disadvantage: It is difficult to generalise any findings because of the sample size and |

| |possible lack of control. |

Glossary Pages 3 and 4

|Investigation |Definition |

|Hypothesis – Directional |A prediction which tells us how the experimenter thinks the IV |

| |will affect the DV. Eg Particpant doing a memory test in a noisy room will score less than a|

| |control group in a quiet room. |

|Hypothesis – Non-directional |A prediction by an experimenter which says that the IV will affect the DV, but doesn’t say |

| |how Eg men and women will score differently on a personality test. |

|Null Hypothesis |A prediction that the IV will have no effect on the DV or that any difference will be due to|

| |chance |

|Experimental Design |Depending on the research, experimenters may choose to use different participants in each |

| |condition (independent measures) or have the same participants do both conditions (repeated |

| |measures) |

|Independent Groups |An experimental design where participants take part in only one condition. This avoids order|

| |effects, but can be affected by participant variables. |

|Repeated Measures |An experimental design where the same participants take part in all conditions of an |

| |experiment. This can lead to order effects, but avoids participant variables |

|Matched Pairs |An experimental design that matches participants for important characteristics such as age, |

| |gender, IQ. Participants do only one condition, which avoids order effects. The matching |

| |process reduces participant variables. |

|Design of Questionnaires |This involves creating a standard list of questions to be given to all participants. The |

| |researcher would decide on how many “Open” or “Closed” questions to ask, depending on |

| |whether they were looking for quantitative or qualitative data. |

|Design of Natural Observations |Researchers decide whether their research will be overt or covert, and then create an |

| |observation schedule or checklist to help them record the behaviour they are interested in. |

| |The may train observers to guarantee inter-observer (or inter-rater) reliability. |

|Design of Interviews |The researcher may choose to use a structured interview (with a standard list of questions |

| |like a questionnaire) or an unstructured interview, where a subject is explored without a |

| |list of questions. Unstructured interviews require a lot more skill and the data they |

| |produce is harder to analyse. |

|Operationalised Variables |The researcher decides what they want to measure when investigating a particular variable. |

| |For example, stress may be operationalised as the increase in a participant’s heart rate. |

|Independent Variable |The variable manipulated by the experimenter to investigate its effect on the DV. |

|Dependent Variable |The variable measured by the experimenter to see if it has been affected by the IV |

|Pilot Studies | A Small-scale study conducted by the researcher before the main study to check whether the |

| |standardised procedures work in an experiment, or to findout if the questions on a |

| |questionnaire are clear. |

| | |

|Extraneous Variable |Any variable other than the IV that might have an effect on the DV. For example, temperature|

| |may have an effect on performance in a memory test where you are comparing males or females.|

|Confounding Variable |Any variable other than the IV that does have an effect on the DV. For example, temperature|

| |may have affected performance in a memory test where you are comparing males or females |

|Reliability |Refers to how consistent a study’s results are. If you repeat the study using standardised |

| |procedures and get similar results, then those findings can be said to be reliable. |

|Test-Retest Reliability |This involves repeating an experiment after a reasonable amount of time and comparing the |

| |results. If they are similar the findings are reliable. |

|Split-Half Reliability |The questions on IQ or personality tests should be consistent. To check, we randomly split |

| |the test into two halves – if the scores are similar, the test is reliable. |

|Inter-Rater reliability |This involves comparing the scores of researchers on an observation. If there is found to be|

| |more than 805 agreement, those scores are reliable. |

|Validity |This refers to how truthful findings from research are. For example, if we have controlled |

| |all extraneous variables in a lab experiment we can be fairly sure that it is true to say |

| |that the IV has caused the change in the DV. This is called internal validity. |

|Experimental Validity |This is where we are sure that - because the experiment has been well-controlled – the IV |

|(Internal Validity) |has caused the change in the DV |

|Ecological validity |We believe that the findings are generalisable outside of the research to the target |

| |population, because the sample was representative and because the experiment had mundane |

| |realism. |

|Content Validity |This is where we are happy that the content of a test or questionnaire is focused on the |

| |area of interest. For example, you wouldn’t include maths questions in an English skills |

| |test. |

|Concurrent Validity |This where participants score the same results on a new test as they would on a |

| |well-established test. For example you would expect people to score similarly on two |

| |different personality tests. |

| | |

|Construct Validity |This is where a test fully covers all the parts of the subject being investigated. For |

| |example, an iQ test would not have construct validity if it only measured maths ability. |

|Ethical Issues |Definitions |

|BPS Code |A list of guidelines, written by the British Psychological Society, that psychologists are |

| |expected to follow when doing research. The code aims to make sure that psychologists |

| |respect the rights of their participants. |

|Ethical Issues – Protection from Harm|Participants should not experience physical or psychological harm when taking part in |

|(Risk of Harm Yr 12) |research. This includes embarrassment or lowered self-esteem. |

|Ethical Issues – Informed Consent ( |Participants must be given comprehensive information about the purpose of research, and |

|Valid consent for Yr 12) |their part in it, so that they can make an informed decision about whether or not to take |

| |part. |

|Ethical Issues – Deception |A participant is not told the true aims of research and so cannot give valid consent. It is |

| |often done to prevent dmand characteristics. |

|Ethical Issues – Right to withdraw |Participants have the right to withdraw form research at any time if they feel uncomfortable|

| |at any point. They also have the right to refuse permission for the researcher to use their |

| |data. Especially important where valid consent hasn’t been possible |

|Ethical Issues – Confidentiality |The participant should not be identified personally within the reporting of research, and |

| |are generally identified by a participant number – they may not wish other people to know |

| |how they behaved in a particular piece of research. |

|Ethical Issues – |The area of our lives that no-one has the right to invade – this might be our personal |

|Privacy |space, or it might be to do with places where we would not expect to be observed. |

|Ethical Issues – |A group of people in a university who decide whether a study can go ahead on ethical |

|Ethical Committees |grounds. They would so a cost-benefit analysis weighing the possible risk to participants |

| |against the possible benefits the research might bring. |

|Ethical Issues |Evaluation and Dealing With issues |

|BPS Code |The code is not a legal document, and so psychologists might be tempted to try and bend the |

| |rules to achieve valid results. However, a serious breach of the code could lead to a |

| |psychologist being expelled from the BPS and unable to do further research. |

|Ethical Issues – Protection from Harm|Psychologist try and do research which does not cause risk of harm greater than the |

|(Risk of Harm) |participant’s everyday life. However, it is very difficult to predict the harm that might be|

| |caused by research. |

|Ethical Issues – Informed |It is difficult to give the participant sufficient information about research without |

|Consent(Valid Consent) |causing demand characteristics – its often dealt with by debriefing the participants after |

| |the study is completed, and explaining the true purpose of the research. |

| | |

|Ethical Issues – Deception |Deception should be approved by an ethics committee that has analysed the costs and benefits|

| |of the research. Debriefing should be used, but it can’t turn the clock back if any |

| |psychological harm has been done. |

|Ethical Issues – Right to withdraw |Though they may feel uncomfortable, participants may remain in the study for two reasons – |

| |if they have been paid expenses, they may feel an obligation to stay. Or, they might simply |

| |not want top soil the research. |

| | |

|Ethical Issues – Confidentiality |Researchers should not record participant names – however in some situations it may be |

| |possible to work out identities from other information eg a school that took part in a |

| |study. |

| | |

|Ethical Issues – |The researcher should not study people without valid consent unless it’s public behaviour in|

|Privacy |a public place. However not everyone agrees what a “public” place is. |

| | |

|Ethical Issues – |Ethical committees often have non-psychologists on them to provide a more balanced decision.|

|Ethical Committees |However, as the majority are psychologists, they may still be biased in favour of the |

| |researcher. |

| | |

|Ethical Issues – |Participants who have been deceived are told the true purpose of the research at the end of |

| |the study. However, you can’t turn the clock back, and some psychological harm may have been|

|Debriefing |done. |

| |Participant Selection |

|Random Sampling | |

|Voluntary Sample | |

|Opportunity Sample | |

|Stratified Sample | |

|Quota Sample | |

|Demand Characteristics | |

|Investigator Effects | |

| | |

|Data |Definition |

|Quantitative Data | |

| | |

|Qualitative Data | |

| | |

|Bar Charts | |

| | |

|Histograms | |

|Scattergraphs | |

|Tables | |

| | |

|Measures of Central Tendency | |

|Mean | |

| | |

|Median | |

| | |

|Mode | |

|Measures of Dispersion – Standard | |

|Deviation | |

|Range | |

|Analysis of Correlational Data | |

|Presentation of Qualitative Data | |

|Content Analysis | |

|Coding System | |

|Categorisation | |

|Descriptive Statistics | |

|Inferential Statistical Test | |

|Levels of Measurement | |

|Nominal Data | |

|Ordinal Data | |

|Interval Data | |

|Ratio Data | |

|Level of Significance | |

|Chi- Squared Test | |

|Mann-Whitney U-Test | |

|Sign Test | |

|Wlcoxon Matched Pairs- Signed Ranks | |

|Test | |

|Spearman’s Rank Order Correlation | |

|Coefficient | |

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download