GoConqr has a new look! Let us know what you think.

Sign up free

Flashcards

research methods booklet 1

megan trent Psychology English 72 cards

Cards in this deck (72)

  1. 1Laboratory ExperimentsAn experiment is a way of conducting research where one variable is changed/manipulated (generally by the experimenter) to see its effects on another variable. The effects on the second variable are measured to see the amount of change.
  2. 2Key points of lab experiment:- IV has been deliberately manipulated by the experimenter.
    -They usually take place in an artificial environment so that the experimenter is able to have control over any extraneous variables which could potentially affect the results.
  3. 3in lab experiments we can clearly see ifour manipulation of the IV is CAUSING the EFFECT seen on the DV. Therefore we can draw conclusions about cause and effect when using Laboratory Experiments.
  4. 4Examples of Laboratory ExperimentsMilgram’s study of Obedience (Social Psychology)
    Asch’s study into majority influence (Social psychology)
    Loftus’s study into the affect misleading information has on the accuracy of EWT (Memory)
  5. 5Evaluation of Laboratory Experiments Strengths:High control over variables - In a lab experiment the majority of the extraneous variables can be controlled. These include: Participant variables, situational variables and experimenter variables. This prevents them becoming confounding variables which distort the results by affecting the DV. This means the experimenter is able to establish cause and effect.
    Easy to replicate - This is because lab experiments tend to follow easily repeated procedures. This means the experiment can be replicated to discover if the results are consistent over time.
  6. 6Evaluation of Laboratory Experiments limitations:Lack Ecological Validity – As Lab experiments take place in an ‘artificial’ environment this means participants may not act naturally in the experiment. This means it is difficult to generalize the results to a real life setting.
    Demand Characteristics – These occur when participants try to make sense of a situation they are in and act accordingly. This might lead the participants to try to guess the aim of the research and, as a result, produce behaviours which they think the research is demanding. This means the results of the experiment may lack validity.
    Experimenter Effects - It is possible that the experimenter/investigator either subconsciously or consciously gives out clues about the true aim of the research which the participants might pick up on. This could be in the way they communicate with the participants either verbally or through their body language. They might also be biased in their recording of data especially if they have an expectation about the results.
  7. 7Ways of controlling EV’s ( Artificiality) overcoming ArtificialityConduct in a natural environment - If participants take part in an experiment in their usual environment they are more likely to behave naturally. This means the results of the experiment are more likely to be valid and accurate.
  8. 8Ways of controlling EV’s (Demand Characteristics) overcoming Demand CharacteristicsDeception - This is means misleading or withholding information from participants about the specific aims of the research. The rationale of this deliberate deception, is that if participants are not told the true aim of the study then they are less likely to be suspicious or curious and change their behaviour
    Use a single-blind technique - This technique is used in experimental research. Suppose researchers wanted to discover whether people prefer the taste of Coke or Pepsi and asked participants to taste each drink in turn. In a single blind condition, the participants would not know in which condition they drank Pepsi or Coke; however, the researcher would know this information.
  9. 9Ways of controlling EV’s (Experimenter Effects)Use a double-blind condition - Similarly to a single-blind condition, this technique is used in experimental research. However, in this technique, neither the participants nor the researcher would be aware which condition participants drank which drink.
    Use standardised instructions - This involves the experimenter using a standardised script which all participants are exposed to. The purpose is to ensure consistency and avoid biases such as leading questions.
  10. 10Field ExperimentsIn a Field experiment generally the IV has been deliberately manipulated by the experimenter. They usually take place in the usual environment of the participants (e.g. a school, a workplace, a hospital, on the street) as they try to increase the realism of the research. There is some control over extraneous variables although less than in a laboratory experiment because it takes place in a more natural setting.
  11. 11Examples of Field ExperimentsHofling’s study into obedience to authority (Social Psychology)
    Bickman’s research investigating the power of uniforms (Social psychology)
    Lorenz’s study into imprinting in no-human animals (Development psychology)
  12. 12Field Experiments AdvantagesImproved Ecological Validity - This is because a field experiment takes place in a participant’s usual environment, such as at work or in their homes. This means their behaviour is more likely to be natural and the results to be valid
    Reduction of Demand Characteristics – In some field experiments participants are unaware they are actually taking part. This means, there is less chance for participants to produce behaviours which they think the research is demanding.
  13. 13Field Experiments WeaknessesDifficult to replicate – It is more difficult to repeat experiments which take place in a real life setting because each situation is different. This means it is difficult to establish if field experiments have external reliability.
    Less control over variables than Lab Experiment – Therefore, extraneous variable may be more prevalent in field experiments as the researcher has less control over variables. This means it is more difficult to establish a cause and effect relationship.
    Ethics – when participants are not aware that they are in an experiment it incurs a lack of informed consent.
  14. 14natural experimentsthe IV has not been deliberately manipulated by the experimenter, it varies naturally. There tends to be very little (if any) control over EVs. They often take advantage of natural settings, situations or behaviours and so tend to be more ecologically valid.
  15. 15quasi experimentsa type of natural experiment where the researcher is unable to freely manipulate the independent variable or randomly allocate the participants to the different conditions. The IV is therefore based on existing differences between people. An example would be gender where males and females are being compared.
  16. 16examples of natural experimentsRutter’s Romanian orphans, the IV being early or late adoption.
    Kiecolt-Glaser’s research into stress and the immune system, the IV being whether before or during stress.
  17. 17examples of Quasi experimentsDay care research where the IV compares those children attending day care with those raised at home.
    Comparison of the attitudes of Psychology and Sociology students towards independent study, the IV being which social science you study.
  18. 18Natural Experiment Evaluations AdvantagesHigh Ecological Validity – This is because a natural experiment usually takes place in a natural environment. This means participant’s behaviour is completely natural and the results can be generalized to a setting outside in which the original study took place.
    No Demand Characteristics – Because participants’ are unaware they are taking part in research, this means there is less chance of them producing behaviour which they think the research is demanding.
    Enables research to be carried out even when there are ethical and practical constraints.
  19. 19Natural Experiment Evaluations WeaknessesVery little control over variables – A natural experiment does not include any involvement or manipulation of variables from the researcher. This means high amounts of extraneous variable are likely, therefore cause and effect relationships cannot be established.
    Very hard to replicate – Due to the lack of researcher intervention, replication is a real issue for natural experiments. This is because each situation and scenario is unique, making it very difficult for the researcher to repeat a study of this nature.
    Sample bias – as participants are not randomly allocated to groups, participant variables may be having the effect rather than the IV.
  20. 20which experiment cannot randomly allocate participants to conditions?Quasi experiments
    may cause confounding variables they can however sometimes be carried out under controlled conditions, sharing the strengths of a lab experiment.
  21. 21Experimental Designsinvolve the ways in which we use our participants in experimental research (Laboratory, field and natural experiments).
  22. 223 experimental designs?Repeated measures
    Independent measures
    Matched pairs
  23. 23Repeated MeasuresIn this design the same participants are exposed to all conditions. For example, if a psychologist wanted to find out the effects of alcohol on a persons driving ability, there would be two conditions. For example
    Condition one Alcohol
    Condition two No alcohol
  24. 24Independent MeasuresFor this design, different participants are used in separate conditions. For example, in an experiment looking at the effects of loud music on memory two separate groups would be exposed to one of two conditions.
  25. 25Matched PairsFor matched pair designs different participants are used in each condition but participants are matched according to an ability or similarity (e.g. IQ or height). For example if we were testing the performance of two new types of golf clubs then we might match our two separate groups of participants for their golfing ability and give each group a different type of club to test.
  26. 26Repeated Measures AdvantagesFewer participants needed - This is because participants in this design taken part in both condition. Therefore, a repeated measures design is likely to less time consuming compared to independent or matched pairs design
    Participant variables kept constant - As participants take part in all condition this means that individual participant variables should be reduced, such as memory, IQ, tolerance to alcohol.
  27. 27Repeated Measures WeaknessesOrder effects - This is because participants take part in more than one condition in this experimental design, which may improve their performance in the second condition. Performance can also suffer due to boredom or fatigue.
    Demand characteristics - Because participants take part in more than one condition they may start to become suspicious and try to guess the aims of the research.
  28. 28Independent Measures AdvantagesNo order effects – This is because participants only take part in one condition. Therefore order effects are not an issue or potential bias in a independent measures design.
    Fewer demand characteristics – As participants only take part in one condition, this means there is less likelihood of participants becoming suspicious and trying to guess the aim of the research.
  29. 29Independent Measures WeaknessesMore participants needed – This is because different participants’ are used in the separate condition; whereas in repeated measures the same participants’ are used in both conditions
    Participant variables - Participant variables relate to the fact that all participants are individuals; all with their own experiences, abilities and personalities. These individual differences could affect the results. Participant variables include things like tolerance to alcohol, IQ, motivation levels.
  30. 30Matched Pairs AdvantagesNo order effects - This is because participants only take part in one condition. Therefore order effects are not an issue or potential bias in a matched pairs design.
    Fewer demand characteristics – As participants only take part in one condition, this means there is less likelihood of participants becoming suspicious and trying to guess the aim of the research
  31. 31Matched Pairs WeaknessesSome participant variables – Even though participants are matched according to abilities, or similarities, each individual is unique and different
    Time consuming – Matching participants according to an ability or similarity is more time consuming compared to the other experimental designs discussed.
  32. 32Ways of Overcoming design flaws:Order effects: Can be overcome by counterbalancing where half the participants take part in condition A followed by condition B and half take part in condition B followed by condition A. eg participant 1: A – B, participant 2: B – A, participant 3: A – B and so on (the ABBA technique!). This does not remove order effects but spreads them equally between conditions so that they do not confound the experiment.
    Participant variables: Can be lessened by random allocation to conditions within an independent measures design. Random allocation evenly distributes participant characteristics across all conditions. An example of a Random technique is to have pieces of paper with A or B written on them placed in a hat which is then shaken and the researcher selects them one at a time to assign participants to groups.
  33. 33NON EXPERIMENTAL RESEARCH METHODS OBSERVATIONAL TECHNIQUESObservations might focus on one particular type of behaviour (e.g. aggression), they might be used to see what category a person falls into (e.g. securely or insecurely attached) or might be used to observe how people behave in certain situations (e.g. on a first date). Depending on what the research is looking at contributes to which observational methods are employed.
  34. 34Naturalistic observationsinvolve observing behaviour in a natural setting. There is no attempt to deliberately manipulate any variables or to influence the environment of the participants being observed so the researcher is not interfering with the behaviour in any way. Usually the participants are completely unaware that they are being observed.
  35. 35Controlled observationsinvolve observing behaviour in prepared situation created by the researcher. There is no attempt to deliberately manipulate any variables or to influence the environment of the participants being observed. So, essentially the researcher is not interfering with the behaviour in any way.
  36. 36Participant vs Non Participant Observations:In participant observations the researcher becomes part of the group of people being observed (e.g. joins a group of football hooligans). In non-participant observations the researcher does not become part of the group being observed and instead just watches the person/group from a distance (e.g. watching a couple on a date in a restaurant)
  37. 37Disclosed (Overt) vs Undisclosed (Covert) Observations:In disclosed (overt) observations the participants are aware that they are being observed, whereas in undisclosed (covert) observations the participants are unaware that they are being observed.
  38. 38Structured vs Unstructured Observations:This related to how the data is collected during the observation. In structured observations data is usually collected in a pre-prepared collection grid and would generally produce quantitative (numerical) data. In an unstructured observation it is likely that the data is collected in a less formal way, for example the research might just make notes on the behaviours which they view as important. This tends to produce more qualitative (descriptive) data.
  39. 39Observational DesignData can be gathered using visual recordings like videos and photographs, audio recordings or note taking using agreed rating scales or coding categories.
  40. 40Behavioural categoriesare defined by operationalising variables so that they can be recorded as objectively as possible during the observation. Behaviour can be recorded using written descriptions or coded eg T = talking. It can also be rated on structured scales eg driving ability on a scale of 1 – 5.
  41. 41Event samplinginvolves counting the number of times a behaviour occurs in a target individual or individuals.
  42. 42Time samplingis when behaviour is counted in a set time frame eg recording the behaviour being exhibited every 30 seconds
  43. 43Pilot StudyIt is important that the researcher conducts a pilot study to practice their observation technique and check inter-observer reliability. This involves checking that the categories are appropriate and ensuring all the observers are recording in the same way.
  44. 44Inter-Observer ReliabilityThis procedure increases the reliability of the findings. It involves the observers agreeing prior to the observation, on the operationalised definition of the behaviour. During the observation the observers independently rate the behaviour. After the observation the observers compare their results using a correlational analysis. If their results positively correlate then it indicates that the data is reliable as they have consistently rated the same behaviours as the behaviour being studied.
  45. 45Naturalistic observations StrengthsIn naturalistic observations there tends to be higher ecological validity, as the participant is being observed in their usual setting. This is even more likely if it is an undisclosed (covert) observation as they are unaware their behaviour is being recorded and so will not alter their behaviour in any way. Therefore the results will not be affected by demand characteristics.
    Observations can be used as preliminary research or in situations where deliberate manipulation of variables would be unethical or impractical, eg football hooliganism or the study of children.
  46. 46Naturalistic observations weaknessIn naturalistic observations there is no control over any extraneous variables therefore cause and effect cannot be determined.
    Observer bias can occur as the observer is having to interpret the behaviour.
    Replication is impossible due to the lack of control over variables, conditions cannot be repeated in exactly the same way.
    A weakness of undisclosed naturalistic observations is that the participants are unaware that they are being observed. As a result this method is potentially breaching ethical issues, such as consent, privacy and deception.
  47. 47questionnaireis a set of written questions designed to collect information from participants on their views and opinions on one or more topics. Questionnaires usually involve a large sample of the target population as they do not require the participant coming into a laboratory, or a researcher having to go out and conduct an observation, instead the questionnaire can be completed at home in the participants’ own time.
  48. 48Open QuestionsOpen questions allow the participant to answer in their own words as their response is not restricted in any way. These types of questions produce qualitative data
  49. 49Closed QuestionsClosed questions provide the participant with prepared potential responses and the participant is asked to select the answer which best represents their view. These types of questions produce quantitative data
  50. 50Questions and instructions must be clear and simpleQuestions and instructions need to be written so that the reader (the respondent) understands what is being asked. It is important that the questions and instructions are clear so that all participants interpret the information in the same way. If participants interpreted the information differently it would affect the reliability of the results.
    Ensure clarity by carefully operationalising (clearly defining) all important terms.
    Ensure that there are no ambiguous questions (questions that can be interpreted in more than one way).
    Ensure the wording of a question doesn’t lead the respondent to be more likely to give a particular answer. This kind of question is called a leading question.
  51. 51Questions should be relevant and ethicalThe completion of a questionnaire should be as simple and concise as possible. The questions included should be directly related to the research study, so no unnecessary questions should be included. Questions should not be too intrusive or invasive.
  52. 52Conduct a Pilot StudySending the questionnaire out to a small group of people from the target population allows the researcher to gather information on any changes which need to be made to the instructions or the questions (or answers for closed questions), prior to the completion of the full scale study. This improves the reliability of the questionnaire.
  53. 53SELF REPORT METHODS: QUESTIONNAIRES strengthsA strength of questionnaires is that they are able to gather a large amount of data relatively quickly.
    If a pilot study has been conducted then it is likely that the procedure is likely to be quite reliable as all partcipants will be exposed to the same set of instructions and the same set of questions to complete. Also the absence of a researcher during the completion of the questionnaire reduces the chances of investigator bias.
  54. 54SELF REPORT METHODS: QUESTIONNAIREs weaknessesSocial desirability bias is a significant problem in questionnaires. This is where participants offer or select an answer which portrays them in a positive light, socially. Therefore the results might not actually reflect the participants’ true opinions and behaviour, meaning that the questionnaire may lack internal validity.
    There may be ethical issues associated with the questions. Participants may feel that their privacy is being invaded if the questionnaire is related to a sensitive topic, or they may feel psychologically harmed by having to answer questions related to distressing events.
  55. 55interviewinvolves the participant verbally answering a series of questions in real time, either face to face or over the telephone.
  56. 56Semi-Structured Interview.generally interviews take the form of the interviewer having a pre-prepared set of questions which they can adapt and refine, dependant on the responses of the interviewee. This type of interview would be regarded as a Semi-Structured Interview.
  57. 57Structured Interview.The format of interviews can vary. Sometimes an interviewer will ask a series of prepared closed questions to the interviewee in order to gain quantitative data; this would be an example of a Structured Interview.
  58. 58Unstructured InterviewSometimes the interviewer doesn’t prepare any questions prior to the interview and therefore the interview follows the format of a documented conversation. This would be considered an Unstructured Interview.
  59. 59Designing Interviews: Important ConsiderationsQuestions should be clear and unambiguous and interviews should avoid leading questions, which might bias the interviewee’s response.
    In interviews another important consideration is how the interviewer may impact on the participant.
  60. 60Behaviour and Appearance of the Interviewer:Behaviour: The way the interviewer/researcher acts towards the participant may affect their responses. For example if an interviewer/researcher was very friendly to one participant but quite abrupt with another then the responses given by each participant might be affected by the way the interviewer/researcher has treated them. The behaviour of the researcher must be consistent across all the interviews.
    Appearance: The appearance of the interviewer/researcher could also affect how comfortable the participant feels. If they are dressed in a lab coat then the participant might feel more intimidated than if the interviewer/researcher is dressed in regular clothing.
  61. 61INTERVIEWS strengthsA semi-structured or unstructured interview is often able to provide very detailed information for the partcipant, because the interviewer is able to carefully clarify the responses given to each question. This means that the researchers interpretation of these responses is likely to be more valid.
    Semi-structured and unstucured interviews may also enable the interviwer to build a rapport with the partcipant, meaning that the participant may feel more comfortable to respond in an open and honest way to the questions asked. This again would increase the validity of the responses and might also provide new areas for investigation.
  62. 62INTERVIEWS weaknessesUsually the data collected from interviews is in the form of long descrpitive qualitative responses. This may mean that analysing the data to gain meaningfull conclusions can be very difficult.
    Another problem with interviews is that they often are time consuming to conduct as the interviewer will need training on how to conduct a successful interview, they take time to actually complete and the analysis is also a time consuming process.
    As the researcher is face to face with the participant during the interview, it is possible that investigator effects may influence the responses given by the participant, due to their tone of voice, reactions to the responses, general appearance etc.
  63. 63Correlational analysisis a way of analysing data. In a correlational analysis we are looking to see if there is a relationship between two variables (or co-variables).
  64. 64Two pieces of information can be obtained from doing a correlational analysis:The direction of any relationship/correlation
    The strength of any relationship/correlation
  65. 65The direction of any relationship/correlationThis can be either POSITIVE or NEGATIVE. A positive correlation means that as one variable increases so does the other variable (for example as the number of hours of revision increases, the scores on psychology unit tests increase). A negative correlation means that as one variable increases the other variable decreases (for example as the amount of alcohol drunk increases the ability to concentrate decreases)
  66. 66The strength of any relationship/correlationThis can be WEAK or STRONG. The strength of the relationship between two variables is measured by a statistical value between -1 and +1. This value is called a CORRELATION COEFFICIENT. The closer the correlation coefficient is to -1 or +1 the stronger the relationship (for example a correlation coefficient of +0.89 would indicate a STRONG positive correlation). The closer the correlation coefficient is to 0 the weaker the relationship (for example a correlation coefficient of -0.26 would indicate a WEAK negative correlation)
  67. 67Presenting Correlational DataScatter graphs are used to represent correlational data. Scores for co-variable 1 are plotted against scores for co-variable 2. You should be able to tell from your scatter graph whether a positive, negative or no correlation exists for your data
  68. 68Hypotheses can beDirectional or Non Directional
  69. 69If you are writing a directional hypothesis for a correlation you will bepredicting that there will be either a positive or a negative correlation between the two co-variables.
  70. 70If you are writing a non-directional hypothesis for a correlation you will bepredicting that there will be a relationship between the two co variables, but does not state the direction of the relationship.
  71. 71Strengths of Correlational AnalysisA strength of correlational analysis is that it allows researchers to undertake exploratory research into a particular area to see whether a relationship exists between two variables. If a relationship is found then further (more experimental) research can be undertaken to see how these two variables are related, whereas if no relationship is found then it is likely a causal relationship can be ruled out.
    Correlational analysis can be one way of investigating an area where other methods might be unethical, as they may make use of existing data to plot the two variables.
  72. 72weaknesses of Correlational AnalysisA weakness of correlational analysis is that we cannot determine cause and effect. This is because a correlation is just looking at whether two variables are related, and is not able to tell which variable has caused the rise or fall in the second variable. It is even possible that a third variable is producing the relationship.
    As with experiments correlations may lack internal validity as it is possible that the measurements of the variables are not accurately measuring the variables (e.g. IQ scores may not be accurate measurements of IQ)

Information

Description
Psychology (Paper 2) Flashcards on research methods booklet 1, created by megan trent on 01/12/2018.
Created by
megan trent
Views
0
Rating

Feedback

What is it about?

What do you think of the new look?

How do you like it?