Methodology

• Steps in Research

  1.     Identify the research problem

  2.     Formulate a hypothesis

  3.     Design the method

  4.     Collect the data

  5.     Analyse the data

  6.     Interpret the results

  7.     Report the findings
• Formulate a Hypothesis

  A statement or testable prediction, about the likely outcome of an experiment by making prediction about the direction of interaction between the IV and DV.

  Population, Independent, Prediction, Dependant, Control.
• Participant Allocation

  Random allocation – ensures every member of the sample has an equal chance of being chosen for the experimental/control group.
• Population and Sample

  Population – the entire group of people belonging to a particular category. The larger group of research interest from which the sample is drawn.

  Sample – group of participants in a research study.

   
• Variables

  Independent variable – condition that an experimenter systematically manipulates, changes or varies in order to gauge its effect on another variable.

  Dependant variable – variable measured in an experiment.

  Controlled variable – any variable kept constant in research conditions
• Extraneous Variables – any variable other than the IV that can impact the results and have an unwanted effect on the experiment.

  Participant differences – individual differences in memory, motivation, mood etc

  Experimenter effect – differences in how the experimenter treats the participants, including non-standardised procedures.

  Artificiality – conducting the experiment in an unnatural environment
• Confounding variables - If extraneous variable are not controlled they can have such a large impact that they have a confounding effect on the interpretation of results. Here the IV causes the change in the DV, rather than the IV.
• Experimental Research Designs

  Between Subjects

  1.     Independent-group design – involves randomly allocating members of the sample to either the control of experiment group. Its quick, and allows for a large number of participants. However groups may differ due to participant differences.
• Experimental Research Designs

  Between Subjects

  2.     Matched participant Design – involves pairing each participant based on a certain characteristic shared, and then randomly allocating one to the experimenter group and the other to the control group. Minimises participant difference but is time consuming, requires a pre-test to match participants, and means that if one in a pair drops out, the other’s results must also be excluded from the study.
• Experimental Research Designs

  Within Subjects

  1.     Repeated Measures Design – using one group of participants and exposing them to both the control and experimental conditions. Eliminates participant differences and allows for fewer participants. However the order effect – a change in results due to the sequence in which two tasks are complete – occurs. This is minimised by counterbalancing where half the group experience the control first, and half experience the experimental conditions. It does not eliminate the order effect, but minimises it.
• Placebo and the Placebo effect

  A placebo is a fake or false treatment used so that the participants aren’t aware whether they are being exposed to the experimental condition. It minimises the placebo effect.
• Placebo and the Placebo effect

  Single Blind

  Where the participants are unaware of whether they are in the experimental or control group, reducing the impact of participant expectations on the results. However, the experiementer effect can occur – where the experimenter’s behaviour towards the groups may influence the results through resulting in the unintentional change in participant behaviours (demand characteristics).
• Placebo and the Placebo effect

  Double blind

  Neither the experimenter nor the participant are aware of which group participants have been allocated to. Removes the participant and experimenter effects but requires an additionally person to allocate the groups who cannot be directly involved with the participants.
• Convenience Sampling

  involves selecting participants based on the researcher’s availability to them.

  Adv:

  • Cheap, quick, convenient

  Dis Adv

  • Not representative due to predisposed quality. Highly biased.
• Random

  Ensures every member of a population has an equal chance of being selected for the sample

  Adv:

  • Quick, cheap, not biased

  Dis Adv:

  • Not representative
• Stratified

  Involves breaking the population into ‘strata’ based on shared characterises. Then participants are selected randomly from the strata, ensuring the appear in the same proportion as in the population

  Adv:

  • Not biased, representative of population

  Dis Adv:

  • Time consuming as require information on population characteristics
• Snowball

  a technique to identify and recruit participants for a study in which existing participants recommend additional potential participants who themselves are observed and asked to nominate others, and so on until a sufficient number of participants are obtained.

  Dis Adv:

  • Biased. Not random and not representative.
• Data collection methodologies

  Observational Studies

  Involves an individual observing another individual or group of people in a natural environment and recording observations about the behaviour they witness. It eliminates artificiality but is subject to observer bias – where the observer sees what they want or expect to see, which may result in a biased interpretation of behaviour.
• Data collection methodologies

  Observational Studies

  1.     Cross-sectional study – investigates a group of people over a period of time with testing at regular intervals to gauge behaviours at different times in their lives.

  2.     Longitudinal study – investigates a selection of different people (age, race…) to explore differences n characteristics at a particular moment in time.
• Data collection methodologies

  Observational Studies

  Correlational study

  Seeks to examine whether a relationship exists between two or more variables without the researcher manipulating anything. It does not infer a casual relationship and no conclusions can be drawn.
• Types of Data

  Qualitative Data

  Def: Data that is often expressed as words. It is descriptive.

  Adv: provides valuable feedback/information about why a participant has given an answer.

  Dis adv:  cannot be statistically analysed.

  Subjective qualitative data – persons values, attitudes and opinions

  e.g self-report, survey, open-ended questions, questionnaires, interviews

  Quantitative Data

  Def: data that is expressed numerically.

  Adv: can be statistically analysed.

  Dis adv: doesn’t provided valuable insight as to why a participant has given an answer.
• Types of Data

  Qualitative Data

  Subjective quantitative – person’s values, attitudes, and opinions

  e.g. Likert scale, close ended questions

  Objective quantitative – how something physiological is measured

  e.g.  blood pressure, heart rate, brain waves, temperature

  Primary Data – empirical evidence. Information psychologists gain from direct observation and measurement.

  Secondary Data – data obtained through secondary sources. E.g archival research, literature review, data modelling.

  Mixed method research – incorporates both qualitative and quantitative measures.
• Correlation – means association/relationship. The extent to which two variables are related which does not infer causes and effect.

  Types of variables in correlation studies:

  -       Behavioural variables – are very difficult to test or manipulate as they are usually unethical. They occur naturally.

  Types of correlation:

  -       Positive – a relationship between two variables, in which both variables move in the same direction. Therefore when one variable increases, the other will increase.
• Correlation

  -       Negative correlation – a relationship between in which an increase in one variable is associated with a decrease in another variable

  -       Zero correlation – no relationship between the two variables.

  Scatter Plots

  Indicates the strength and direction of the correlation between the co-variables. It doesn’t matter which variable goes on which axis.

  Correlation coefficient - When working with continuous variables, the correlation coefficient to use if Pearson’s R.
• Ethical guidelines – participant rights

  1.     Voluntary participation

  a.     Participant willingly decides to take part.

  b.     Participants must not experience pressure, or threats or be coerced to participate.

  2.     Informed consent

  a.     The researcher must obtain written permission from each participant stating they consent and that they have been given all necessary information.

  b.     Inform the participant of their rights and any possible harm (physical or psychological)

  c.     If the participant in underage, a parent/guardian should complete the form.
• Ethical guidelines – participant rights

  3.     Withdrawal rights

  a.     Participants can cease participation at any time without negative consequences or pressure to continue.

  b.     Participants may withdraw their results following the conclusion of the experiment.

  4.     Confidentiality

  a.     Participant’s right to privacy in the access, storage and disposal collected about them.

  b.     Participant involvement and results cannot be disclosed without written consent.
• Ethical guidelines – participant rights

  5.     Deception

  a.     Used when giving the participants information beforehand may influence their behaviour and affect results.

  b.     Must be used with caution and researchers must ensure all participants are thoroughly debriefed.
• Ethical guidelines – participant rights

  6.     Debriefing

  a.     Participants are informed of the study’s true purpose.

  b.     Correct any mistaken attitudes or beliefs held by participants, and explain all deception related to the conducting of the experiment.

  c.     Provide an opportunity for the participant to gain access to results, procedures and conclusion, and provide additional support through counselling as required.
• Validity – whether a test actually measures the effect of the ID on the DV.

  Content Validity – means the contents of the test, including all subtests and items adequately measure what it is design to measure.

  Criterion-related validity – means the test can adequately predict performance on other tasks.

  Concurrent validity – comparing results with other comparable scores.
• Construct validity – means the test provides a good reflection of the theory on which it is based and that there is empirical evidence supporting the theory.

  Internal validity – refers to the design of the research and procedures used.

  External validity – means the conclusions can be generalised to the population from which the sample was drawn.
• Reliability – refers to the ability of a test to produce consistent results.

  Test-retest Reliability – conducting the experiment/test twice to compare the scores. One problem is the ‘test practice effect’.

  Parallel-forms Reliability – involves giving another version of the same test rather than the same test twice.

  Split-half reliability – involves dividing the original test into halves and examining the correlation between scores on each half. Establishes internal reliability.
• Internal consistency – using correlations between different items in the same test to determine whether the items produce similar scores.

  Inter-rater reliability – involves checking that different test administrators get similar results from it.
• Ethical Concepts

  Refers to the moral principles and codes of behaviour that apply to all psychologists, regardless of the field in which they work.

  Prior to the commencement of any research, the researcher submits a research plan to an ethics committee for approval. The ethics committee ensure the study follows the principles:

  1.     Beneficence – weighing the potential risk (physical and psychological) against the potential benefits to society.

  2.     Non-maleficence – avoiding causing any physical or psychological harm. Implies that harm/discomfort is not outweighed by benefits.
• Ethical Concepts

  3.     Respect – consideration of an individual’s welfare and an appreciation of their own uniqueness, autonomy and freedom of expression

  4.     Justice – ensures fair treatment for all

  5.     Integrity – the researchers commitment to the honest conducting and reporting of research
• Errors and Uncertainty

  Inferential statistics – allow a researcher to make inferences about the results of an experiment, to form conclusions and generalise findings of the population. To do this they explore whether the data obtained are a result of manipulation of the IV in the experiment or whether the results are due to chance, subject to error or contain reasonable amount of uncertainty.
• Errors and Uncertainty

  Errors:

  -       Personal – errors made by the researcher such as an experiment

  -       Random – errors that account for unpredictable variation or impact on results

  -       Systematic – errors that occur in a consistent manner in relation to the true data value which can be the result of something like a confounding varible.

  Statistical significance – refers to the significance of the difference between scores – whether the difference demonstrates that the results are attributed to the IV and not chance. When it is strong, there is little uncertainty among the results.
• Animals in Research

  1.     Replacement: permit a given purpose of an activity or project to be achieved without the use of animals

  2.     Reduction: methods for obtaining comparable levels of information with few animals in scientific procedures or for obtaining more information from the same number of animals

  3.     Refinement: methods that alleviate or minimise potential pain and distress and enhance animal wellbeing

  Remember findings in experiments regarding animals cannot be generalised to humans
• Scientific and non-scientific research

  The scientific method involves a systematic approach to the planning, conducting and reporting of research, usually in the form of experiments. With a scientific approach, researchers collect empirical evidence or data the enables them to draw on valid conclusions based on experiments. The scientific method is based on, concerned with or verifiable by observation or experience.

  Non-scientific means it is not based on science and do not have any scientific foundation – pseudoscience. No verifying observations happening.
• Experimental or non-experimental

  Experiments are conducted to test whether one variable influences another variable – they test cause and effect relationships.

  Features of experimental research methods:

  1.     Independent and Dependant variable

  2.     Directional Hypothesis

  3.     Controlled and uncontrolled variables

  4.     Experimental and control groups

  5.     Placebo and Experimenter effects

  6.     Reliability and Validity

  7.     Longitudinal and Cross-sectional designs
• Non-experimental methods in psychological research often involve observation, the collection of data by carefully watching and recording the behaviours of others. Related to qualitative methods and the variables cannot be manipulated by the experimenter.

  1.     unmodifiable qualities -age, gender

  2.     cannot be modified for ethical reasons

  3.     not feasible to modify