Research Methods

Features of Psychology as a Science

The process of scientific study:

1. Observation of subjects

2. Formulation of a theory

3. Construction of a testable hypothesis

4. Empirical testing of the hypothesis

5. Replication and validation

6. Falsifiability

• Empirical- Empirical data is information gained through direct observation rather than reasoned argument or belief 

• Objective- Data should not be affected by the expectations of the researcher. Data collection should be systematic and free from bias. Without objectivity there is no way of knowing if our findings are valid.

• Controlled- All extraneous variables need to be controlled in order to be able to establish cause (IV) and effect (DV).

• Replication- scientists record their methods and standardise them carefully so that the same procedures can be followed in the future (replicated). Repeating a study is the most important way to demonstrate the validity of an observation or experiment. If the outcome is the same, then this indicates that the original findings are valid.

• Hypothesis testing- a statement is made at the beginning of an investigation that serves as a prediction and is derived from a theory. There are different types of hypotheses (null and alternative), which need to be stated in a form that can be tested (i.e. operationalized and unambiguous).

• Predictability- We should be aiming to be able to predict future behavior from the findings of our research.

• The philosopher Thomas Kuhn suggested what distinguishes scientific disciplines from non-scientific disciplines is a shared set of assumptions and methods known as a paradigm. He argued that Psychology lacked a universally accepted paradigm and was therefore best seen as pre-science. Natural sciences on the other hand have a number of principles at their core. Eg biology has the theory of evolution. By contrast, psychology has many conflicting approaches.

• Paradigm shift- occurs when paradigms are challenged more and more over time until there is so much evidence that the old paradigm can no longer be accepted such that new paradigms will take its place. In psychology there was a paradigm shift from the psychoanalysis of the psychodynamic approach to the behaviourist approach and then again to the cognitive approach and then to the biological approach.

Reliability

Reliability- Whether something is consistent. In the case of a study, whether it is replicable.

Types of reliability:

1. Internal reliability assesses the consistency of results across items within a test.

This can be assessed using:

• Split half method- measures the extent to which all parts of the test contribute equally to what is being measured

2. External reliability refers to the extent to which a measure varies from one use to another

This can be assessed using:.

• Inter-rater reliability- extent to which two or more observers are observing and recording behaviour in the same way. We can assess it by comparing the results of two researchers on the same thing and see how well they correlate. We can improve inter rater reliability with training, practice in a pilot study, better operationalisation, and ensuring that each researcher has the same ability to see the subjects.

• Test-retest reliability- Involves presenting the same participants with the same test or questionnaire on two separate occasions and seeing whether there is a positive correlation between the two. If correlation is poor then we should alter the task until it produces higher correlation.

Validity

Validity- Whether something is true (measures what it sets out to measure).

Types of validity:

1. Internal validity- The extent to which a test is consistent within itself, such as whether the different questions (known as ‘items’) in a questionnaire are all measuring the same thing.:

• Factors that affect internal validity include (definitions found under Control of extraneous variables):

• Participant variables (i.e: personality, demand characteristics

• Lack of experimental control (this covers order and Investigator effects)

• Situational variables (i.e: time of day and temperature)

• Researcher bias (i.e: lack of objectivity)

Types of internal validity include:

• Content validity- The extent to which the questions/measurements in the study measure what we think we are measuring rather than some other factor.

• Face validity- Simple way of assessing whether a test measures what it claims to measure which is concerned with face value – e.g. does an IQ test look like it tests intelligence.

• Split-half method- Comparing two halves of a test, questionnaire, or interview. We can test this by having participants sit the two halves of the task separately and seeing if their scores correlate (i.e: each person scores similarly in both tasks). To improve this we can remove and/or change questions to improve correlation.

2. External validity- Whether it is possible to generalise the results beyond the experimental setting (i.e: onto the wider population).

Types of external validity include:

• Ecological validity- The extent to which the findings of a research study are able to be generalized to real-life settings

• Mundane realism- is the task similar to those encountered in real life

• Population validity- Whether the research can be generalised to other people/populations.

• Temporal validity- Refers to how likely it is that the time period when a study was conducted has influenced the findings and whether they can be generalised to other periods in time

• Concurrent validity- Comparing a new test with another test of the same thing to see if they produce similar results. If they do then the new test has concurrent validity

External validity is improved by replicating the study in new situations or groups, and via use of field studies and naturalistic observations.

Aims

• Aim- The researcher’s area of interest – what they are looking at (e.g. to investigate helping behaviour).

• Differs from a hypothesis as it is a statement of what you believe to be true rather than a prediction 

Hypotheses

• A hypothesis (plural hypotheses) is a precise, testable statement of what the researchers predict will be the outcome of the study.

• This usually involves proposing a possible relationship between two variables: the independent variable (what the researcher changes) and the dependent variable (what the research measures).

• In research, there is a convention that the hypothesis is written in two forms, the null hypothesis, and the alternative hypothesis (called the experimental hypothesis when the method of investigation is an experiment). We can only accept one, and must reject the other.

• Alternative Hypothesis- The alternative hypothesis states that there is a relationship between the two variables being studied (one variable has an effect on the other). It states that the results are not due to chance and that they are significant in terms of supporting the theory being investigated.

• Null Hypothesis- The null hypothesis states that there is no relationship between the two variables being studied (one variable does not affect the other). It states results are due to chance and are not significant in terms of supporting the idea being investigated.

• Nondirectional Hypothesis- A two-tailed non-directional hypothesis predicts that the independent variable will have an effect on the dependent variable, but the direction of the effect is not specified. Used when there is no previous research (theories/studies) suggesting that findings will go in a particular direction. E.g., there will be a difference in how many numbers are correctly recalled by children and adults.

• Directional Hypothesis- A one-tailed directional hypothesis predicts the nature of the effect of the independent variable on the dependent variable. Used when there is previous research (theories/studies) suggesting that findings will go in a particular direction. E.g., adults will correctly recall more words than children.

• Upon analysis of the results, an alternative hypothesis can be rejected or supported, but it can never be proven to be correct.

• We must avoid any reference to results proving a theory as this implies 100% certainty, and there is always a chance that evidence may exist which could refute a theory.

How to Write a Hypothesis

1. To write the alternative and null hypotheses for an investigation, you need to identify the key variables in the study. The independent variable is manipulated by the researcher and the dependent variable is the outcome which is measured.

2. Operationalise the variables being investigated. Operational variables (or operationalizing definitions) refer to how you will define and measure a specific variable as it is used in your study.

3. Decide on a direction for your prediction. If there is evidence in the literature to support a specific effect on the independent variable on the dependent variable, write a directional (one-tailed) hypothesis. If there are limited or ambiguous findings in the literature regarding the effect of the independent variable on the dependent variable, write a non-directional (two-tailed) hypothesis.

4. Write your hypothesis. A good hypothesis is short (i.e. concise) and comprises clear and simple language.

Sampling

• Representative sample- A sample that that closely matched the target population as a whole in terms of key variables and characteristics

• Target population- The group that the researchers draws the sample from and wants to be able to generalise the findings to

• Identifying a smaller group within this population is called a sampling frame. The sampling frame should be representative of the target population so that generalisations can be made.

• Random sampling- A sampling technique where everyone in the target population has an equal chance of being selected (refers to the practice of using chance methods to allocate participants to the conditions of an investigation)

• Stratified sample- A sampling technique where groups of participants are selected in proportion to their frequency in the target population

• Systematic sample- A sampling technique where every nth person in a list of the target population is selected

• Opportunity sample- A sampling technique where participants are chosen because they are easily available

• Volunteer sample- A sampling technique where participants put themselves forward to take part in research, often by answering an advertisement

Pilot studies

Pilot study- A small scale study conducted to ensure the method will work according to plan. If it doesn’t then amendments can be made. 

Evaluation

• A pilot study can help the researcher identify problems with the design, methodology, and analysis of your study and gain feedback so that these issues are dealt with (study adapted) before you carry out the main/real study. Sometimes the task is too hard, and the researcher may get a floor effect, because none of the participants can score at all or can complete the task – all performances are low. The opposite effect is a ceiling effect, when the task is so easy that all achieve virtually full marks or top performances and are “hitting the ceiling”.

• Requires time and money

• Participants used in the pilot can’t be used in the real study, reducing the number of individuals from your target population available to you

Experimental design

Independent groups design- An experimental design where each participants only takes part in one condition of the IV

Evaluation

• Order effects cannot be observed, as no participants will be used in more than one condition.

• Data collection will be less time-consuming if all conditions of the experiment can be conducted simultaneously.

• Different participants need to be recruited for each condition, which can be difficult and expensive.

• There is a risk of participant variables (individual differences between participants) affecting the results between conditions, rather than solely manipulation of the independent variable.

Repeated measures design- An experimental design where each participants takes part in both/all conditions of the IV

Evaluation

• The results will not be subject to participant variables (i.e. individual differences between participants), putting more confidence in dependent variable changes being solely due to manipulated changes in the independent variable.

• As the same participants are used [at least] twice, extra participants do not need to be recruited.

• There is risk of observing order effects (e.g. practice / fatigue effects, or demand characteristics), but this risk be reduced by counterbalancing (i.e. controlling the order of variables so that each order combination occurs the same number of times, e.g. one half of participants partake in condition A followed by B, whereas the other half partake in B followed by A).

• If a participant drops out, data will be lost from all conditions of the experiment rather than one.

Matched pairs design- An experimental design where pairs of participants are matched on important characteristics and one member allocated to each condition of the IV

Evaluation

• Order effects will not be observed as participants only take part in one condition.

• The tailored participant-matching process reduces the risk of participant variables (individual differences) from affecting results between conditions.

• Different participants need to be recruited for each condition, which is difficult and expensive.

• Matching is a more complex process, and it will always be very difficult to match participants identically.

Variables

• Operationalising variables- This means clearly describing the variables (IV and DV) in terms of how they will be manipulated (IV) or measured (DV).

• Independent variable- The variable that the experimenter manipulates (changes).

• Dependent variable- The variable that is measured to tell you the outcome.

• Extraneous variable- Variables that if not controlled may affect the DV and provide a false impression that an IV has produced changes when it hasn’t.

• Confounding variable- An extraneous variable that varies systematically with the IV so we cannot be sure of the true source of the change to the DV

• For covariables see correlations

Control of Extraneous variable

• Order effects- Order effects can occur in a repeated measures design and refers to how the positioning of tasks influences the outcome e.g. practice effect or boredom effect on second task. These can be controlled using counterbalancing (A way of trying to control for order effects in a repeated measures design, e.g. half the participants do condition A followed by B and the other half do B followed by A)

• Participant variables- participants in one group may differ in a significant way from participants in another group. This risk can be reduced via random allocation (participants randomly assigned to groups) or by matched pairs (see experimental designs).

• Situational variables- factors in the environment that may affect the DV. These can be reduced by using a standard procedure (see experiments).

• Investigator effects- These result from the effects of a researcher’s behaviour and characteristics on an investigation. Can be controlled using double blind control (participants are not told the true purpose of the research and the experimenter is also blind to at least some aspects of the research design) or standardised instructions and/or procedure (see lab experiments).

• Demand characteristics- Occur when participants try to make sense of the research situation they are in and try to guess the purpose of the research or try to present themselves in a good way. Can be reduced using Double blind control (see above) or Single blind control (Participants are not told the true purpose of the research).

Ethics (P.DDI WAC)

• Ethical issues- There are 3 main ethical issues that occur in psychological research – deception, lack of informed consent and lack of protection of participants.

General areas of ethical consideration for when doing research include:

• Privacy – Ensuring that participants are aware that they do not have to answer anything that may make them feel uncomfortable, thus protecting their privacy. In addition, you cannot observe people in private environments i.e. peeping through someone's window, without their knowledge!

• Deception – Participants should not be deliberately lied to about the aim of the study and procedure. If participants are not told the true aim of the study, every step should be taken to ensure that there are no harmful effects to the participant. For example, a thorough debrief, counselling sessions, ethics committee there at all times to stop experiment if they feel that it is harming participants etc.

• Debrief/Brief – After completing the research, the true aim is revealed to the participant. Aim of debriefing = to return the person to the state s/he was in before they took part. We do this as briefing the participant (telling them what will happen in the study before they agree to take part) may lead to demand characteristics

• Informed consent – Psychologists should ensure that all participants are helped to understand fully all aspects of the research before they agree (give consent) to take part

• Withdrawal (right to withdraw) – Participants should be aware that they can leave the study (or remove their data from the study) at any time, even if they have been paid to take part.

• Avoid harming participants mentally/physically or psychologically (protection from harm) – Participants should be protected from physical or mental harm, including stress - risk of harm must be no greater than that to which they are exposed in everyday life, and must leave in the same emotional state that they were in when they entered

• Confidentiality – Participants results and personal information should be kept safely and not released to anyone outside of the study.

Methods for avoiding demand characteristics:

• Prior general consent- Before participants are recruited they are asked whether they are prepared to take part in research where they might be deceived about the true purpose

• Presumptive consent- Asking a group of people from the same target population as the sample whether they would agree to take part in such a study, if yes then presume the sample would

• Retrospective consent- Once the true nature of the research has been revealed, participants should be given the right to withdraw their data if they are not happy.

• Cost-benefit analysis- benefits to society are compared to potential costs to participants (adding to human knowledge vs. harm). If the benefits outweigh the risks then the study goes ahead

• Ethics committee- A group decide whether your research should go ahead using cost-benefit analysis.

Ethical guidelines- These are provided by the BPS - they are the ‘rules’ by which all psychologists should operate, including those carrying out research.

• Responsibility- Valuing the dignity and worth of all individuals. This links to informed consent, right to withdraw and confidentiality.

• Respect- Valuing the responsibilities of being a psychologist. This links to protection of participants and the role of a debrief

• Integrity- Valuing honesty, accuracy, clarity and fairness. This links to the particular importance of maintaining high standards when deceiving participants

• Competence- Valuing the continued development as a psychologist and the maintenance of high standards of work.

Briefing and debriefing are also types of consent, but briefing can lead to demand characteristics and debriefing is less ethical as it requires deception.

Experiments

Laboratory experiment- An experiment that takes place in a controlled environment where the experimenter manipulates the IV and measures the DV. This can be done using:

1. Standardised instructions- The instructions given to each participant are kept identical – to help prevent experimenter bias.

2. Standardised procedures- In every step of the research all the participants are treated in exactly the same way and so all have the same experience.

• Experiments often have a control group (A group that is treated normally and gives us a measure of how people behave when they are not exposed to the experimental treatment), and always have at least one experimental group (The group that received the experimental treatment).

Evaluation

• It is easier to replicate (i.e. copy) a laboratory experiment. This is because a standardized procedure is used.

• They allow for precise control of extraneous and independent variables. This allows a cause and effect relationship to be established.

• The artificiality of the setting may produce unnatural behavior that does not reflect real life, i.e. low ecological validity. This means it would not be possible to generalize the findings to a real life setting.

• Demand characteristics or experimenter effects may bias the results and become confounding variables.

Field experiment- An experiment that takes place in a natural setting where the experimenter manipulates the IV and measures the DV

Evaluation

• Behavior in a field experiment is more likely to reflect real life because of its natural setting, i.e. higher ecological validity than a lab experiment.

• There is less likelihood of demand characteristics affecting the results, as participants may not know they are being studied. This occurs when the study is covert.

• There is less control over extraneous variables that might bias the results. This makes it difficult for another researcher to replicate the study in exactly the same way.

Natural experiment-  These are conducted in the everyday (i.e. real life) environment of the participants, but here the experimenter has no control over the IV as it occurs naturally in real life

Evaluation

• Behavior in a natural experiment is more likely to reflect real life because of its natural setting, i.e. very high ecological validity.

• There is less likelihood of demand characteristics affecting the results, as participants may not know they are being studied.

• Can be used in situations in which it would be ethically unacceptable to manipulate the independent variable, e.g. researching stress.

• They may be more expensive and time consuming than lab experiments.

• There is no control over extraneous variables that might bias the results. This makes it difficult for another researcher to replicate the study in exactly the same way.

Quasi experiments- those where the change in IV cannot be manipulated or randomly assigned (i.e: male/female, young/old).

Evaluation

• Due to the IV naturally occurring within the individual it may be more reflective to that individual.

• They allow researchers to investigate variables that would be unethical to manipulate

• Control over extraneous variables is often difficult. As the researcher is not manipulating the IV, they can be less sure that it caused an EFFECT on the DV.

• They are generally hard to replicate and therefore can lack internal and external reliability.

Observations

Naturalistic observation- An observation study conducted in the environment where the behaviour would normally occur
Evaluation

• By being able to observe the flow of behavior in its own setting studies have greater ecological validity. 

• Like case studies naturalistic observation is often used to generate new ideas. Because it gives the researcher the opportunity to study the total situation it often suggests avenues of enquiry not thought of before.

• These observations are often conducted on a micro (small) scale and may lack a representative sample (biased in relation to age, gender, social class or ethnicity). This may result in the findings lacking the ability to be generalized to wider society.

• Natural observations are less reliable as other variables cannot be controlled. This makes it difficult for another researcher to repeat the study in exactly the same way.

• A further disadvantage is that the researcher needs to be trained to be able to recognise aspects of a situation that are psychologically significant and worth further attention. 

• With observations we do not have manipulations of variables (or control over extraneous variables) which means cause and effect relationships cannot be established.

Controlled observation- An observation study where the researchers control some variables - often takes place in laboratory setting

Evaluation

• Controlled observations can be easily replicated by other researchers by using the same observation schedule. This means it is easy to test for reliability.

• The data obtained from structured observations is easier and quicker to analyze as it is quantitative (i.e. numerical) - making this a less time consuming method compared to naturalistic observations. 

• Controlled observations are fairly quick to conduct which means that many observations can take place within a short amount of time. This means a large sample can be obtained resulting in the findings being representative and having the ability to be generalized to a large population.. 

• Controlled observations can lack validity due to the Hawthorne effect/demand characteristics. When participants know they are being watched they may act differently.

Overt observation- Also known as a disclosed observation as the participants given their permission for their behaviour to be observed

Evaluation

• Reduces ethical issues as the participants are aware that they are being observed.

• If observed for long periods of time, people tend to forget about observers and behave more naturally, particularly when being filmed.

• Increase of social desirability as they are aware of being observed, they may change their behaviour in a way that they feel is favourable to others, which decreases validity.

• Increase chance of demand characteristics, as the participant is aware of the researcher, they may change their behaviour in order to fit in with what they think the researcher wants to see.

Covert observation- Also known as an undisclosed observation as the participants do not know their behaviour is being observed

Evaluation

• Increases validity as participants aren’t aware of being observed, they are more likely to act naturally.

• Less demand characteristics as the participant isn’t aware of the researcher, there would be less chance of them changing their behaviour to fit in with the researchers expectations.

• It creates ethical issues as the participant has not consented to being observed. This makes it difficult to follow other ethics such as right to withdraw, debrief etc.

• If the participant becomes aware of the researchers presence, they may change their behaviour, thus decreasing validity.

Participant observation- Observation study where the researcher actually joins the group or takes part in the situation they are observing.

Evaluation

• Only way to observe some behaviours i.e. cults/gangs

• Greater accuracy and detail as the participant is involved in the behaviour. This allows the researcher to make more valid conclusions about behaviour as they have a greater insight and perspective

• It can be difficult to get time / privacy for recording. For example, with covert observations researchers can’t take notes openly as this would blow their cover. This means they have to wait until they are alone and rely on their memory. This is a problem as they may forget details and are unlikely to remember direct quotations. 

• If the researcher becomes too involved they may lose objectivity and become biased. There is always the danger that we will “see” what we expect (or want) to see. This is a problem as they could selectively report information instead of noting everything they observe. Thus reducing the validity of their data.

• The presence of the researcher can influence behaviour, reducing validity

Non-participant observation- Observation study where the researcher does not join the group or take part in the situation they are observing.

Evaluation

• Easier to remain objective as the observer is away from distractions and can remain focused – increasing validity.

• The researcher has less influence on behaviour therefore chances of researcher effect are reduced. The participant is likely to be less influenced by the researcher.

• If participants are unaware that they are being observed as part of a non-participant observation, it can raise ethical issues such as consent and the right to withdraw.

• Less detail and accuracy as the researcher is at a distance from the participants, therefore, some behaviours may be interpreted or recorded inaccurately – reducing validity.

There are two types of non-participant observation:

• Structured observation- An observation study using predetermined coding scheme to record the participants' behaviour. This produces quantitative data

• Unstructured observation- Observation where there is no checklist so every behaviour seen is written down in an much detail as possible (often produces qualitative data)

Evaluation of structured

• Easier to record as there is a specific focus on certain behaviours. This increases the validity, as the researcher is not distracted by other behaviours, or behaviours that may be irrelevant to the research aim.

• Easier to establish inter-rater reliability. Due to the clear, planned focus on behaviour, the research could be easily used and understood in a consistent way, also improving replicability.

• Can reduce validity as there is a clear focus, behaviours that may be important may be missed due to it not being part of the planned behaviours.

• Open to observer bias as the researcher may interpret behaviours in a way that fits into the planned behavioural categories, therefore reducing validity as it may not reflect what actually happened.

Evaluation of unstructured

• Increases validity as the researcher is taking into account all behaviours that are going on. This ensures that more valid conclusions are made of behaviour, as a wide perspective is gathered, not a small focused one.

• Applicable to a wide range of contexts. This is a key strength as this method is extremely easy to use when collecting data on many different situations of behaviour.

• Harder to record as the researcher has to pay attention to everything around them, this may cause behaviours to be missed and reduce validity.

• Harder to establish reliability because there is limited focus, therefore if it were to be replicated, the focus is likely to be different in relation to the individual carrying out the observation.

• Open to observer bias as the researcher may only note down behaviours that support their own theories, or behaviours that reflect what they hoped to find.

Recording of Data

The two main coding schemes for observations are:

1. Event sampling- A target behaviour is identified and the observer records it every time it occurs

Evaluation

• Records are easy to obtain and analyse as researchers can clearly see the total number of behaviours for each event. This can make analysis extremely quick and easy, especially when looking for most or least common behaviours.

• More reliable observations as the events are already planned, therefore it could be easily replicated to measure consistency of observational behaviours.

• Can miss important behaviours due to having set events already planned, other behaviours that were not considered are missed – reducing validity.

• If many events occur at once it may lead to behaviours not being recorded – reducing validity.

• It gives no indication of the amount of time spent on each behavioural category, therefore it can sometimes lead to less valid conclusions about behaviour.

2. Time sampling- A way of sampling the behaviour that is being observed by recording what happens in a series of fixed time intervals.

Evaluation

• Less likely to miss behaviours as the researcher usually has a short time to focus on recording behaviour, therefore is more likely to be accurate.

• It can give an indication of how much time is spent on each behaviour.

• Behaviours that occur outside the time intervals are not accounted for, therefore may reduce validity as important behaviours may be missed.

• Can be hard if lots of behaviour occurs at once

• Can miss events not coded for – reducing validity

Types of time sampling:

Instantaneous scan sampling: The action performed at the start of each preset interval is recorded. For example, if an observer were watching one child’s aggressive behaviour in the playground, they might record at the start of every 10 second interval i.e. 10, 20, 30, 40 seconds, whether he or she displayed 1.Aggression 2.Non-aggression 3. Interacting with others. They would ignore the child’s activity at any other time.

Predominant activity sampling: The same time periods and categories can be used, but instead the researcher watches throughout the whole interval and only records the behaviour that the individual performed the most during that time.

One-zero sampling: The researcher can use the same time intervals as outlined above, but instead, they would record whether the behaviours occurred or not in the time period. For example, going back to the child aggression study, within the 10 seconds i the researcher could have seen all three behaviours (aggression, non-aggression, interaction with others), and therefore would tick all of them.

To record behaviours we use behavioural categories. Behavioural categories contain a list of key behaviours, or collections of behaviour, that the researcher conducting the observation will pay attention to and record.

• Coding Frames – Allow for more specific behaviours to be observed within a behaviour category. Codes and abbreviations can be used to record the severity of behaviours or a different subtype within a category.

• Example of Behavioural Category – Kicking, Punching

• Example Codes – Kung Fu Kick – KFK,

• Severity of strength of the Kicks–K1,K2,K3,K4,K5

Inter-rater reliability for observations

Researchers observing the same behaviour and coding the behaviour in the same way. If there is low inter-rater reliability it suggests:

1. the coding of behaviour is either vague or lacks validity

2. they’re not observing the same event.

Evaluation of observations in general

• See how people behave rather than how they say they behave.

• Allows us to study variables it would be unethical to manipulate e.g. behaviour in prisons.

• Useful as pilot to generate hypothesis for future research.

• Difficult to replicate

• Does not provide us with thoughts or feelings, only behaviour

• No manipulating variables, so can’t establish cause and effect

• Observer bias- Occurs when the observers know the aims of the study or the hypotheses and allow this knowledge to influence their observations

• Evaluation apprehension- Participants’ behaviour is distorted as they fear being judged by observers

• Time consuming and requires careful preparation

To increase validity of observations:

• Carry out a covert observation so participants don’t change their behaviour (observer effect)

• Double blind observations to reduce observer bias

• Clearly operationalised coding system.

To increase reliability of observations:

• Clearly operationalised coding system

• Check inter-rater reliability

• Train researchers to use coding system to ensure there is a consistent understanding of the behavioural categories

• Conduct a pilot study to check behaviour categories

Self-report methods

Self-Report- Asking a participant about their thoughts and behaviour and recording their answers.

Self-reports can be used in a variety of different ways:

• Questionnaires

• Diary entries

• Interviews

• Psychometric tests

They can be used with or without manipulation (so in an experiment or on their own).

Validity can be improved by:

• Removing leading/unclear/poorly operationalised/socially desirable/recall questions (can find which ones these are using a pilot study)

• Adding open questions with qualitative data.
  Adding closed questions to allow for easy quantitative data collection

• Add filler questions unrelated to the study so that participants don’t figure out the aim and display demand characteristics

• Ensuring answers will be anonymous and confident.

Reliability can be improved by:

• Training interviewers so they are standardised.

• Providing standardised questions.

• Adding closed questions with quantifiable data.

• Using split test/test-retest methods.

Self-reports: Interviews

Structured interview- Interview where the questions are fixed and the interviewer reads them out and records the responses

Evaluation

• The structured nature means that the interview can easily be repeated to increase internal reliability

• Structured interviews can be easily assessed for reliability and improved by removing or changing inconsistent items

• Interviewer doesn’t require as much training as for semi-structured or unstructured interviews

• Structured interviews are limited by fixed questions, lack validity.

• Harder to build rapport with participant, so may not get as much or as high quality information

Semi-structured interview- Interview that has some predetermined questions, but the interviewer can develop others in response to answers given by the participant

Evaluation

• Enables the researcher respond more flexibly and so gain more detailed information than from a structured interviews

• Answers to the set questions allow for analysis

• Difficult to assess for reliability as questions asked can alter, difficult to repeat exactly as questions alter, so may lack internal validity

Unstructured interview- Also known as a clinical interview, there are no fixed questions just general aims and it is more like a conversation

Evaluation

• May provide higher level of rapport, making it easier to gain knowledge from the participant

• Enables the researcher respond more flexibly and so gain more detailed information than from a structured interviews

• Requires highly trained interviewer to stay on track

• Only some people willing to participate, so not representative of the population

• Participants may be affected by biases such as social desirability or leading questions

• Greater risk of low internal validity and demand characteristics having more of an effect as difficult to replicate the interview exactly due to differing questions each time

Evaluation of interviews in general

• Can generate quantitative data if questions are closed and qualitative if they are open

• Researcher bias can occur. The expectations of the interviewer may alter the way the respondent answers questions.

• Subjectivity (questions responses can be open to interpretation depending on intonation and body language used and who witnesses them).

• Social desirability bias- Participants’ behaviour is distorted as they modify this in order to be seen in a positive light.

Self-reports: Questionnaires

Questionnaire- A set of written questions that participants fill in themselves

Qualitative data- Descriptive information that is expressed in words

Quantitative data- Information that can be measured and written down with numbers.

Closed questions- Questions where there are fixed choices of responses e.g. yes/no. They generate quantitative data

Types of closed question

Fixed choice questions are phrased so that the respondent has to make a fixed choice answer usually ‘yes’ or ‘no’.

Do you revise?      Yes/No

Checklist questions give a list of options and told to choose as many as apply to the participant:

Which of these methods do you use to revise? (Tick all that apply.)

Mind maps

Mnemonics

Story technique

Use of imagery

Flashcards

Quizzes

Ranking questions are whereby participants are instructed to put a list of options into order.

Rank the following activities according to how much time you spend on them each day

(1 = most time, 5= least time):

Revising

Homework

Reading

Texting

Using social media sites

Likert scale questions are whereby participants indicate on a scale how much you agree with a statement. It is also known as a verbal rating scale.

“Psychology is the most important subject ever”. Circle one answer.

Strongly agree

Agree

Unsure

Disagree

Strongly Disagree

Rating scale questions: This type of question asks the the individual a question, and then they must highlight on a numerical scale where they feel best reflects their view

Semantic differential questions indicate where you stand on a scale between two contrasting adjectives, such as; good/bad, hungry/full, and exciting/boring.

Psychometric Questionnaires: A series of standardised closed questions, to measure a mental characteristic such as IQ, emotional intelligence and personality traits.

Evaluation of closed questions

• Closed questions are quick and easy for participants to answer.

• Closed questions are more likely to be structured in a certain order, therefore high in internal reliability.

• Due to time efficiency, large samples can be collected increasing generalisability.

• Quantitative data easy to analyse e.g. find median, modes and draw graphs.

• Lacks detail, participants can’t express opinions fully, lacks validity.

• Risk of response bias e.g. saying yes to everything.

• The score for all participants on each question is only nominal data so only the mode can be calculated. Limited analysis.

Open questions- Questions where there is no fixed response and participants can give any answer they like. They generate qualitative data.

Evaluation of open questions

• They produce qualitative data, giving participants an opportunity to fully express their opinions, thus increasing validity.

• All info is analysed so information is not lost by averaging answers – increasing validity.

• Qualitative data is time consuming to analyse as themes need to be identified.

• Interpretation of data is subjective, leading to bias. This can lead to issues of validity. In addition, the inconsistency of interpreting data can lead to low inter-rater reliability.

• Findings are based on individuals so may lack generalisability.

Evaluation of questionnaires

• Questionnaires are a relatively cheap and quick way to gather a large amount of data.

• Since questionnaires can be completed privately (and often anonymously), responses may be more likely to be honest. However, not having an experimenter to supervise its completion could present a problem.

• Social desirability issues may arise, where participants give incorrect responses to try to put themselves in a socially acceptable light. Can be solved by adding in easy filler questions.

• Distributing questionnaires en masse (e.g. via post or the internet) means that any data collected relies on responses to be returned; response rates are often poor, plus it may be that only a certain type of person returns questionnaires, so generalising the sample of results to a large population can be unconvincing.

• Questionnaires may be flawed if some questions are leading (i.e. they suggest a desired response in the way they are worded).

• If any questions are misunderstood, participants completing questionnaires privately cannot get clarification on the meaning/responding accurately from an experimenter, so may complete them incorrectly.

Case studies

Case study- In-depth investigation of a single person, group or event, where data is gathered from a variety of sources and by using several different methods (e.g. observations & interviews).

Evaluation

• Provides detailed (rich qualitative) information.

• Provides insight for further research.

• Permitting investigation of otherwise impractical (or unethical) situations.

• Can’t generalize the results to the wider population.

• Researchers' own subjective feeling may influence the case study (researcher bias).

• Difficult to replicate.

• Time consuming.

Content analysis

Thematic analysis- A method for analysing qualitative data which involves identifying, analysing and reporting patterns within the data

Content analysis- Technique used to analyse qualitative data which involves coding the written data into categories – converting qualitative data into quantitative data.

• Content analysis is a research tool used to indirectly observe the presence of certain words, images or concepts within the media (e.g. advertisements, books films etc.). For example, content analysis could be used to study sex-role stereotyping.

•  Researchers quantify (i.e. count) and analyze (i.e. examine) the presence, meanings and relationships of words and concepts, then make inferences about the messages within the media, the writer(s), the audience, and even the culture and time of which these are a part.

•  To conduct a content analysis on any such media, the media is coded or broken down, into manageable categories on a variety of levels - word, word sense, phrase, sentence, or theme - and then examined.

Evaluation

• High ecological validity as based on real interactions

• Easy to get sample as not reliant on participants

• Easy to replicate

• It is a reliable way to analyse qualitative data as the coding units are not open to interpretation and so are applied in the same way over time and with different researchers

• It is an easy technique to use and is not too time consuming

• It allows a statistical analysis to be conducted if required as there is usually quantitative data as a result of the procedure

• Causality cannot be established as it merely describes the data

• As it only describes the data it cannot extract any deeper meaning or explanation for the data patterns arising.

• Risk of observer bias (i.e: fitting observations into what they want to find).

Correlation

• Correlational analysis- A mathematical technique where the researcher looks to see whether scores for two covariables are related

• Co-variables- The variables investigated in a correlation (correlations have these instead of an IV and a DV)

• Scattergram- Used to plot correlations where each pair of values is plotted against each other to see if there is a relationship between them.

• Positive correlation- A relationship exists between two covariables where as one increases, so does the other

• Negative correlation- A relationship exists between two covariables where as one increases, the other decreases

• No correlation- No relationship exists between variables

• The Correlation Coefficient is a number between -1 and 1 that tells us how strong the relationship is.

• +1.0 perfect positive correlation

• +0.8 strong positive correlation

• +0.5 moderate positive correlation

• +0.3 weak positive correlation

• 0 no correlation

• -0.3 weak negative correlation

• -0.5 moderate negative correlation

• -0.8 strong negative correlation

• -1.0 perfect negative correlation

• When writing hypotheses for correlations, use the word relationship when talking about the two variables. Null= no significant relationship between, One-tailed= there will be significant negative/positive a relationship between, Two-tailed= their will be a significant relationship between

Evaluation

• Makes a good pilot study to generate a hypothesis for an experiment.

• Can research variables that would be unethical to manipulate.

• Can understand the relationship between two variables (positive/negative, weak/strong).

• Correlations do not show causation.

• They have the same weakness as whatever method was used to gather the data for the co-variables (i.e: observation/self report).

• Correlations can’t show/highlight other variables that may be the real cause

• Often correlations are misleading i.e. Bacon causes cancer!!!! does it?

Meta-analysis

Meta-analysis- A statistical technique used to gather data from lots of studies on the same topic, and combine them to see the overall effect.

Evaluation

• Large number of participants overall, so more likely to have high population validity

• The results can be saved from being skewed through using the effect size which is the size of the effect observed in each study. Looking at the effect size of all the studies and comparing them then it can be observed whether they are reliable or not because if they all have similar effect sizes then they support each others findings and the overall statistical analysis. 

• Using Meta-analysis is also a cheaper way to conduct your research as you do not have to pay expenses for carrying out an experiment. 

• The results can provide strong evidence for the hypothesis and also become more generalisable. 

• It can also gain data from different types of studies such as analogue and clinical studies so that they can be compared.

• Meta-analysis is more susceptible to researcher bias than other research methods. The researcher collecting the data may pick specific studies that only provide the outcome that the researcher is looking for. 

• Also, not all details of a study that is used may be given as the the researcher collecting the data who was not actively involved in the study at the time it was done and so some data may be missing which can affect the overall result of that study and reduces its reliability. 

• Poor designs are also mixed with good ones which can skew the statistical result.

Psychology and the Economy

• The economy concerns the production, distribution, and consumption of goods and services.

• The implications of psychological research for the economy are concerned with how the knowledge and understanding gained from psychological research (theories and studies) may contribute towards our economic prosperity. 

• Mental health: if more effective treatments can be developed for psychological health problems then this means that people will be able to return to work and this reduces the burden on the employers, NHS and taxpayer. 

• Social influence: If effective ways of influencing people to change their behaviours can be found, then people can be prevented from drinking and smoking, reducing their risk of ill health and thus death such that they are more likely to be able to work, allowing them to contribute to the economic prosperity of the country. 

• Criminal psychology: Better eyewitness testimony and cognitive interviews lead to a more efficient justice system. 

• Childhood development: understanding attachment allows us to create and apply interventions to children at risk of insecure attachment so that they can develop into productive adults.

Peer review

Peer review- Before going to publication, a research report is sent other psychologists who are knowledgeable in the research topic for them to review the study, and check for any problems

Questions that a peer reviewer would consider:

• Introduction- 1. Is the research appropriate for their study?​

2. Is the research suggested original and interesting?​

3. Is the current state of the area of research, accurately represented?​

4. What specific questions does the study address?​

• Approach/method- 1. Were appropriate controls used?​

2. Was is ethical?​

3. Can you think of a better way to address the research question?​

• Results- 1. Do the figures/tables/data used contribute to the paper or are they useless?​

2. Are the authors interpretations of the results backed up by the data?​

3. Are the statistical analyses appropriate?

The peer reviewer, after answering the above questions must decide to either:

• Accept the paper ​

• Reject the paper ​

• Suggest amendments

Evaluation

• Bias – a reviewer may strongly support an opposing view, making them less likely to provide an unbiased opinion of the work. Many believe that it is not possible to separate a reviewer from their personal, political or cultural values. Similarly, the reviewer is more likely to look favourably upon research presented by someone within their social circle. Other forms of bias include Institution bias (research from prestigious universities is favoured) and Gender Bias (male researchers seem to be favoured)

• Can be hard to find a suitable expert

• Anonymity. Researcher is not supposed to know the identity of the peer reviewer, however this can allow the peer reviewer (as they know the identity of the researcher) to abuse their power and use their peer review to manipulate the researcher (i.e: settle old scores) so can lack objectivity.

• The ‘file-drawer’ problem – there is a bias towards publishing studies with positive results i.e., those supporting the hypothesis. Negative findings tend to be either rejected or are never submitted for publication. For every study showing positive findings, there could be a hundred with negative results stuffed in university filing drawer – our understanding of a subject then becomes distorted.

• Can preserve the status quo and prevent new ideas being explored as peer review tends to favour research that supports existing theories rather than original or dissenting work.

• Ensures that only research with sound methodology is passed, stopping poorly conducted research being accepted as a fact by the public, meaning that only valid research is accessed by the public.

• Allows us to assess the research ratings of universities. All university departments are expected to produce research, which is then assessed by peer review, and funding can be allocated based on this.

• Issues of bias can be overcome by using a double-blind peer review. This is when both the reviewer and the author are anonymous, and therefore no extra information is provided which could influence their judgement.​

Reporting Psychological studies

• Psychological reports allow psychologists to showcase their findings of important and ground breaking research. It is important for Psychologists to detail all of their ingredients used from materials used, to results. Another reason why Psychologists publish research is so that other psychologists can replicate the study conducted. A report contains the following sections:

1. Title

2. Abstract

3. Introduction

4. Aims and hypotheses

5. Method 

6. Design

7. Results

8. Discussion

9. References

10. Appendices

• The Title should say what the study is about and include the IV and DV.

• The Abstract should be a concise summary of the study, with a brief description of the aims, hypothesis, method, and a summary of the results. Usually between 150 and 200 words long.

• The Introduction starts with general theory, briefly introducing the topic, then narrows down to specific and relevant theory and research (2 or 3 studies).  The aim and alternate and null hypotheses are stated after this.

• The Method describes how the research was carried out. Someone should be able to replicate the study by following the method, so it needs to be detailed. The method should include sufficient detail so that other researchers are able to precisely replicate the study; Design- the design is clearly stated, eg independent groups, naturalistic observation etc, and the reasons/justification given for the choice. Sample- information related to the people involved in the study; how many there were, biographical/demographic information (as long as this does not compromise anonymity) and the sampling method and target population. Apparatus/materials- detail of any assessment instruments used and other relevant materials. Procedure- a 'recipe-style' list of everything that happened in the investigation from beginning to end. This includes a verbatim record of everything that was said to participants; briefing, standardised instructions and debriefing.

• Ethics- an explanation of how these were addressed

• Results- An explanation of the findings – summarizing the results and relating them to the aim and hypothesis. The null hypothesis should be accepted or rejected in the discussion. Any unexpected findings should also be addressed and explained here. The relationship to background research – the results need to be related to the background research and covered in the introduction. The data should be compared to other data and comments made on whether or not the findings of the study support the findings of other studies. The implications of the study – for example, whether the study relates to real life situations, e.g. interviews, exams etc. The limitations and modifications of the study – any problems or limitations need to be explained, along with modifications that could improve the study. Suggestions for further research – at least two ideas for further research should be included.

• Discussion- This contains a summary of the results, relating them to the aims and hypothesis, and also considering the methodology and implications of the study and making suggestions for future research.

• References- The references section contains a list of all the books, articles and websites that have been used for information during the study. It allows the reader to see where the information on the research and theories mentioned in the report (e.g. the introduction) came from. References should be presented in alphabetical order of first author’s surname.

• Appendices- Appendices should contain the following items:

• Consent Form– a copy of the one you give to participants.

• Standardised Instructions– a script like document which illustrates the experiences of all participants in each experimental condition.

• Ethics Sheet– this will appear in your consent sheet which will outline the ethical issues that will not be breached.

• Materials– these could be, for example, the word lists used in memory experiment, powerpoint slides etc.

• Raw Data– these should be in the form of tables illustrating both condition results obtained for each participant.

• Statistical Analysis– any calculations conducted on data should appear here.

How to write a reference

To write a Harvard reference for a book:

• Put the surname then initials for each author (et al not used), the the title of the book in italics, then the place of publication, then the publisher

To write a Harvard reference for a journal article:

• Put the surname the initials for each researcher, then the year of publication (in brackets), then the title of the study, then the journal title (in italics), the volume number (in bold), the issue number of the journal, then page number(s)

Quantitative and Qualitative data

• Quantitative data is numerical (in the form of numbers) and is gathered by collecting individual scores from participants. It is often gained from large scale experiments or questionnaires.

Evaluation

• Can be analysed statistically and converted into graphs and summarised

• Precise

• A reliable measure as it is objective

• Measures behaviours

• Lacks detail

• Often collected from artificial situations so may lack validity

• Qualitative data is non-numerical, so often in the form of words. It often comes from interview transcripts, unstructured observations, and case studies.

Evaluation

• Rich and detailed

• Often collected in a real-life setting so has high external validity

• Low reliability as subjective

• Imprecise

However it can be converted into Quantitative data via content analysis and behavioural categories

Primary data

• Primary data- Information that the researcher has collected firsthand for a specific purpose e.g. data from an experiment or observation

Evaluation

• Is gathered first hand, therefore there is more certainty on how valid it is, as the researcher themselves knows the strengths and weaknesses of their own research.

• If collected objectively, with careful planning and sampling, controls in place and other features of methodology adhered to, then likely to be scientifically gathered for the stated aim of the study. This means they are more credible.

• New research and ideas can be discovered through primary data, as it may not have been explored before.

• Expensive to obtain because each researcher or research team has to start from the beginning of a study and follow the whole study through, finding participants, organising materials and running the study.

• Time consuming, due to the above.

• Limited to the time, place and number of participants etc., whereas secondary data can come from different sources to give more range and detail.

• Maybe biased due to the researcher wanting to find certain results.

Secondary data

• Secondary data- when the researcher uses information collected by others (often to answer other questions) e.g. the work of other psychologists or government statistics

Evaluation

• Doesn’t take long to collect as the research has already been carried out.

• Can gather lots of data in a short space of time.

• Can help to build an idea about what most research is presenting in certain areas

• You don’t always know where it has come from and how reliable it is

• Might not be relevant to what you are researching. This can lead to spending lots of time trawling through journals and research papers.

• Data can be over complicated and may be more difficult to understand. This is because the research has been written by someone else.

• Sometimes the data can be out of date (lacks temporal validity)

Descriptive statistics

• Descriptive statistics- Analysis of data that helps describe, show or summarize data in a meaningful way

• Measures of central tendency- A measurement of data that indicates where the middle of the information lies e.g. mean, median or mode

• Mean- Measure of central tendency calculated by adding all the scores in a set of data together and dividing by the total number of scores

Evaluation of mean

• Most informative as it takes every score into account

• Any data that is greatly larger or smaller (extreme values) in comparison with the other pieces of data can distort the mean

• Sometimes the mean doesn’t make sense in terms of what the data is about e.g. the mean number of children in a family = 2.4

• Median- Measure of central tendency calculated by arranging scores in a set of data from lowest to highest and finding the middle score

Evaluation

• It is less affected by extreme scores

• It is not suited to being used with small sets of data, especially if it contains widely varying scores e.g. 7, 8, 9, 102, 121 = 9, but a more accurate median would be 60!

• Mode- Measure of central tendency which is the most frequently occurring score in a set of data. When there is more than one number that appears the most frequently, we call this bimodal.

Evaluation

• Is not affected by extreme scores

• Gives a good idea of how often something is occurring e.g. what mobile phone is selling the most

• A set of data may not have a most frequent score

• Dispersion measure- A dispersion measure shows how a set of data is spread out, examples are the range and the standard deviation

• Range- The distance between the lowest and the highest value in a set of scores. It is also a measure of dispersion which involves subtracting the lowest score from the highest score in a set of data

Evaluation

• Easy to calculate

• Only using two scores in the data set and ignoring the rest

• The extreme scores could distort the range

• Standard deviation- A measure of the average spread of scores around the mean. The greater the standard deviation the more spread out the scores are.

To work out standard deviation: 1. Work out the Mean (the simple average of the numbers), 2. Then for each number subtract the Mean and square the result, 3. Then work out the mean of those squared differences, 4. Take the square root of that and we are done

Evaluation

• SD is the most sensitive measure of dispersion as it is derived by using every score in the data set 

• Is not very distorted by extreme scores.

• The SD is closely related to the mean and is the best measure of dispersion to use when the mean is being used as the measure of central tendency

• Takes a long period of time to calculate

• Assumes a normal distribution

Inferential statistical tests

• Inferential statistics- Inferential statistics are methods of analyzing data using statistical tests that allow the researcher to make conclusions about whether a hypothesis was supported by the results.

• Tests of difference are used for experiments to compare two conditions and see if there is a difference between them. Related data is when the two sets of data are from the same participants (repeated measures or matched pairs), and unrelated data is from two different sets of participants (independent measures).

• We can also have tests of correlation where we measure two variables and see if there is a relationship. 

• The three kinds of data that inferential statistics cover are nominal data, ordinal data, and interval/ratio data (for definitions see levels of measurement).

• So do No Order Imogen down side vertically, Unis Rate Cats across the top horizontally, then across the rows Carrots Should Come, Mashed With Swede, Under Roast Potatoes to fill the boxes

• For the Sign test and Wilcoxon (See willy) require the critical value to equal to or less than the observed value (bc), and all the other require the critical value to be more than or equal to the observed value.

Chi squared

Calculating degrees of freedom (df)= (r-1) x (c-1)

r= number of rows, and c= number of columns

Sign test

Sign test- A statistical test used to analyse the direction of differences of scores between the same or matched pairs of subjects under two experimental conditions

• We use this test when differences predicted between two sets of related data (such as in an experiment).  N= number of participants, S= the observed value

1. Subtract each value in the control condition from the experimental condition, recording the sign (+ or -) 

2. Count the number of times the less frequent sign appears to give you S

3. Count the total number of pluses and minuses to get N

4. Use the table of critical values for the sign test, picking one or two tailed as appropriate. The critical value at the intersection between N and the 0.05 level of significance. S must be less than or equal to the critical value to be significant

5. State the conclusion (i.e: as the result is significant/not significant we cannot accept the null/alternative hypothesis, therefore we accept the alternative/null hypothesis)

Parametric test criteria

Parametric tests

There are three criteria that must be met in order to use a parametric test;

1. Data must be interval level- parametric tests use the actual scores rather than ranked data.

2. The data should be drawn from a population which would be expected to show a normal distribution for the variable being measured. Variables that would produce a skewed distribution are not appropriate for parametric tests.

3. There should be homogeneity of variance- the set of scores in each condition should have similar dispersion or spread. One way of determining variance is by comparing the standard deviations in each condition; if they are similar, a parametric test may be used. In a related design it is generally assumed that the two groups of scores have a similar spread.

Graphs and tables

• Tables clearly present data and show patterns. Raw data tables contain information/scores directly collected from participants before it is analysed. Raw data can then be summarised in smaller tables which show descriptive statistics such as mean, median, and mode. Tables can also be used to summarise descriptive statistics.

• Frequency tables show you how often a particular event occurs using a tally.

• A contingency table shows you the sum of all value in each row and/or column.

• Bar chart- A graph that shows the data in the form of categories (e.g. behaviours observed, so non-continuous data) that the researcher wishes to compare. The bars should not touch or else it is a histogram not a bar chart.

• Histogram- A graph that is used for continuous data (e.g. test scores). There should be no space between the bars, because the data is continuous. Each column shows a class interval (i.e: people aged 1-16 years).

• Line graph- A graph used for continuous data, and can be used to compare two or more different sets of continuous data.

• Scatter graphs- these are used for correlations (see correlations)

When drawing a graph remember to include:

• A title that includes the IV and DV (i.e: a graph to show the relationship between ...etc). If you have already been given the hypothesis then use it in your title.

• A suitable scale (not huge or tiny given the amount/type of data)

• Correctly labelled axis

• Correctly plotted data

Distribution

• Normal distribution- An arrangement of a data that is symmetrical and forms a bell shaped pattern where the mean, median, and mode all fall in the centre at the highest peak. Examples of data like this includes IQ, shoe size, ad weight in a normal population.

• Skewed distribution- An arrangement of data that is not symmetrical as data is clustered to one end of the distribution. A positive skew is where the data is clustered to the right which means there are few high scores. A negative skew means that data is clustered to the left so there are few low scores.

Probability and significance

Probability- How likely something is to happen – can be expressed as a number (0.5) or a percentage (50% chance of tossing coin and getting a head). In terms of statistical tests near 1 means very likely, and near 0 means very unlikely. If the probability of your results being due to chance is low then you accept your hypothesis as your results are statistically significant and vice versa.

• Critical value- The value that a test statistic must reach in order for the hypothesis to be accepted.

• Observed value- The value that you have obtained from conducting your statistical test

• Significance- If the result of a statistical test is significant it is highly unlikely to have occurred by chance

• Type 1 error- Is a false positive. It is where you accept the alternative/experimental hypothesis when it is false

• Type 2 error- Is a false negative. It is where you accept the null hypothesis when it is false

Levels of measurement

• Nominal level data- Frequency count data that consists of the number of participants falling into categories. (e.g. 7 people passed their driving test first time, 6 didn’t).

• Ordinal level data- Data that is capable of being put into rank order (e.g. places in a beauty contest, or ratings for attractiveness).

• Interval level data- Data measured in fixed units with equal distance between points on the scale