Research Methods


Aims and hypotheses

Aim- A general statement of what the researchers is investigating: the purpose the study

Hypotheses- A statements that state the relationship between the variables investigated

Directional (one-tailed) hypothesis- States the outcome of what's expected from the experiment.

Non-directional (two-tailed) hypothesis- States there will a difference in performance between the two conditions.

Null hypothesis- States there is no differeence between the variables.

Variables- Anything that can change in the investigation.

Independent variable- The variable that is manipulated by the researcher.

Dependent variable- The variable that is measured by the researcher.

1 of 44

Control of variables

Extraneous variable- Any other variables expect from the IV  that could influence the measurement of the DV.

There are three key considerations to take when controlling extraneous variables:

Participant variables – minimising differences between participants (e.g. their stage of development such as age, or ability such as IQ).

Researcher variables – factors such as researcher behaviour, appearance or gender could affect participant responses, so should be made consistent throughout the experiment.

Situational variables – control of the setting where the experiment takes place, such as keeping light, sound and temperature levels consistent.

Confounding variables- If a variable has actually been found to have influenced the results of the research then it would be considered to have confounded the results.

2 of 44

Types of experimental design and control

Independent group design- Different groups of participants are used in each condition of the IV. For example, Group A does a task with the TV on, while group B does a task with the TV turned off.

Random allocation-Independent group design involves random allocation because it ensures that each participant has an equal chance of being assigned to each group.

Repeated group design- The same participants take part in each condition. For example, each participant does a task with the TV on. A week later the participants do the same task with the TV off.

Counterbalancing- Order effects can occur in a repeated measure design. This can be overcome by using counterbalancing. For example, half of the participants do condition A then B, and the other half do condition B then A. This can reduce order effects.

Matched pairs design- Different groups take part in each condition by are matched on key characteristics. For example, one member of the matched pair is allocated to group A and the other group B. 

3 of 44

Experimental design-Evaluation

Independent groups

There are no order effects because participants are taking part in each condition.                     Participants are less likely to guess the aim of the study and change their behaviour.

The groups contain different people (e.g age, gender and background). The differences between the participants may affect the results.

Repeated groups

There are fewer participants are needed to take part in all conditions and this saves time.             The same participants are used in the all the conditions which mean individuals differences are reduced.                                                                                                                               All participants take part in the same condition more than once so the participants may be affected by order effects.

4 of 44

Experimental design-Evaluation

Matched groups                                                                                                            Order effects are reduced as the participants have been matched into their groups.                     It's very time consuming to find closely matched pairs. The researcher may use a large group of participants to obtain matched pairs on key variables.

Dealing with limitations of matched pairs- Each participant should be randomly assigned to their conditions. Also, the researchers should conduct a pilot study as this may be important for matching pairs.

5 of 44

Types of experiments: Lab and Field

Laboratory experiments- They are experiments that take place in a controlled environment where the IV is manipulated and the DV is recorded.                                                                                                                                                                                                         There control as the researcher is able to manipulate the variables. Therefore, they are high in internal validity as they can observe the change of the DV due to the IV. This makes replication easier.

 The participants know they are being studied so demand characteristics may influence the behaviour and this can affect the results. This causes them to be low in validity. 

Field experiment- They are experiments that take place in participant natural environments. Participants are randomly allocated to their conditions.                                                             There is less likelihood of demand characteristics affecting the results because participants are unaware they are being studied. This improves internal validity.

The participants are unaware they are being studied so informed consent may be difficult to get.

6 of 44

Types of experiments: Natural and Quasi

Natural/ Quasi-experiments- These experiments take place in laboratories or natural settings. The IV is being controlled by someone else rather than the researcher.                                        

There is a less likelihood of demand characteristics affecting the results as the participants don't know they are being studied.

There is a lack of control changing the IV means there is a chance of confounding variables influencing the results.

due to the researcher’s lack of control, research procedures cannot be repeated so that the reliability of results cannot be checked.

7 of 44

Sampling Types and Evaluation

Sampling- a method used to select participants

Random sampling- Everyone in the target population has an equal chance of being selected.

There is no researcher bias as all the participants have an equal chance of being selected.

It's time-consuming as the researcher will need a list of all members of the population and contact those selected which is time-consuming.

Stratified sampling- Participants are obtained by identifying subgroups (e.g boys and girls and age groups).                                                                                                                    

The sample is representative of the target population because the participants are randomly selected. This shows the results can be generalised. 

It's very time-consuming to identify subgroups and randomly select participants and contact them.

8 of 44

Sampling Types and Evaluation

Volunteer sampling- The sample consists of people who respond to a request. For example advertising in a newspaper or the internet.                                                                          

Gives access to a variety of participants. This makes the sample more representative and unbiased.  Those who respond to the call for volunteers may all display similar characteristics (such as being more trusting or being more willing to volunteer than those who did not apply) thus increasing the chances of an unrepresentative sample.

Opportunity sampling- A sample is obtained on whoever happens to be available at the time where the study is happening.

It's the easiest method because it uses the first suitable participants the researcher can find. This makes it less time-consuming and easy to obtain a sample. 

The sample may not representative because the sample is obtained from a specific area.

Systematic sampling- Every nth person of the target population is selected. A random sample can be used. The list order has been randomised, this method offers an unbiased chance of gaining a representative sample. Members of the population don't have an equal chance of being selected.

9 of 44

Ethical issues

Ethical issues are problems that arise between the rights of the participants and the goals of the researcher.

The BPS code of ethics is a document provided by the British Psychological Society. It instructs psychologists in the UK on what behaviour is and is not expectable when dealing with participants.

Informed consent is where participants are informed of the true aims of the study or at least telling participants what is going to happen. This helps participants in whether or not they should take part in the study.

Deception is where the researcher is upholding or misleading participants. Therefore, the participants cannot give their consent.

Protection from harm- the participants are harmed by participating in the research. This includes being put under stress, embarrassment, hurt or exposing them to anything that changes their mental or physical state.

Privacy and confidentiality- the researcher must make sure names are hidden and data is anonymous. 

10 of 44

Dealing with ethical issues

Informed consent- informed consent should be dealt with a consent form. For investigations, children who are under the age of 16 informed consent is needed.

Deception- Deception is dealt with by using debriefing by informing participants what the investigation is aiming to do and allow them to ask questions about the study. The right to withdrawal should be given.

Protection from harm- The study should be stopped if harm is suspected. Participants should be given the right to withdraw at any time.

Confidentiality- Researchers should not record the names of the participants instead they should use numbers and false names. 

11 of 44

Observational techniques and Evaluation

Natural observation is where behaviour is studied in a natural setting, where behaviour normally occurs.For example, Schaffer and Emerson studied infant's behaviour in their homes.      

The behaviour is natural meaning there is high ecological validity because the participants know they are being observed.

Cannot be replicated to check reliability, as the researcher is not in control of variables.Cannot be replicated to check reliability, as the researcher is not in control of variables.

Controlled observations are when the researcher is watching and recording the participant's behaviour in a structured environment.                                                                                 

Controlled observations are quick which means observations take place within a short amount of time. This means there is a large sample used which means the findings are representative.

Participants know they are able studied so the results will be affected by demand characteristics

12 of 44

Observational techniques and Evaluation

Covert observations- the participants do not know they are being observed.                           

Participants are unaware they are being studied so behaviour is more natural.

Less ethical as participants are not aware they are taking part and cannot give fully informed consent.

Overt observations- the participants are aware they are being studied.              

It is possible to inform participants in advance and obtain informed consent                  

Participants know they are being studied so they may change their behaviour which affects the validity of the data.

13 of 44

Observational techniques and Evaluation

Participant observation- the researcher becomes a member of the group they are watching 

It gives the researcher increased insight into the lives of those being studied so they are unlikely to overlook or miss behaviours.         

The researcher may become too involved in the study which could cause the participants to guess the aims of the study

Non-participant observation- the researcher is watching the participants from outside the group.                                                                                                                         

Investigator effects are less likely to occur as the researcher is  not visible                                                     

Due to a lack of proximity the researcher might overlook or miss behaviours of interest.

14 of 44

Self report techniques-Questionnaires

Questionnaires are a set of written questions. It is designed to collect information about a topic or topics.

Questions can be:                                                                                                            

Open questions-respondents can provide their own answers. Provide rich detailed answers and unexpected answers.                                                                                                         

Closed questions- respondents are given a range of answers and they can only select one. The answers are easier to analyse by using graphs and measures like mean.

They are given to large samples cheaply and quickly. This enables the researcher to collect large amounts of data which mean that data can be generalised    Respondents feel more willing to give personal information than in an interview.                                                                           

The respondents may not have been truthful in their answers (due to social desirability). This causes the data to lack in validity.                                                                                      Questionnaires are only given to people who can read and write and they will take more time to filling them out. This means that the sample is biased.

15 of 44

Self report techniques-Interviews

Interviews is a research method that is done face to face. It involves interaction with another individual. Interviews can be:                                                                                              Structured- the research has a set of planned questions. The interview asks questions and the interviewee replies.

Structured interviews (as well as questionnaires) can be repeated as the questions are planned in advance. This means the answers from different people can be compared.                                                                                                                          

The interviewer's expectations may influence the answers the interviewee gives. Therefore invesgations effects affect the validity of the responses

Unstructured- They have less little structure. A schedule is not used and they may not be timed. Questions may be added or missed during the interview process.                                 Unstructured interviews allow more detailed information to be collected from the respondents than in a structured interview. This because the interviewer changes the questions which allow the researcher to gain deeper insight to the respondents feelings and thoughts.            The interviewer's expectations may influence the answers the interviewee gives. Therefore investigator effects affect the validity of the responses Social desirability can occur

16 of 44

Self report techniques-Interviews (Evaluation)

Interviews provide the researcher with a good depth of information as the researcher is giving the researher can accurate picture of what they are saying. This allows the researcher to collect detailed responses.

Interviews are time-consuming as the participants is usually interviewd on their own. This means there is a small sample size making the results less generaliseable.

17 of 44

Observational design

Observational research involves wathcing and recording behaviour that is relevant to the research aim. Some research use methods to obtain infomation such as behavioural categories and data extraction such as event and time samplying.

Behavioural categories-Before starting the observation the research decideds what behaviour is relevant to the researcher question.  A tally chart is set up to record it. This consists of different categories of target behaviour and the researcher ticks it when they see it occuring. For example, observing infant behaviour, would includ things such as smiling, crying and sleeping.

It makes data collect more structured and objective for the researcher

It allows the researcher to make sure the catorgories don't overlap

The researcher is able to convert possible behaviour to the researcher aim

18 of 44

Observational design

Event samplying-Counting the times a certain behaviour occurs for exmaple counting the number of times a person smiles in 10 mintues.

Event smaplying is useful when the behaviour happens frequently.               

It can be difficult to achieve as there is alot of action to record.

Time samplying-Recording behaviours in a given time frame for example noting what the individual has done very 30 secs           

Its effective as it reduces the number of observations.    

The researcher can miss out important events because they happen outside the time frame.

19 of 44

Correlations (Evaluation)

Correlations is way to look for the relationship between two variables. Correlations are conducted in secondary data and they are useful for psychological reseach.

Types of correlations:

Negetive correlation is where the variables decrease and the other dercrease.            

Positive correlation the variables increase and so do the others.                              

No correlation is where there is no relationship between the variables.

Correlations are conducted qucikly by using secondary data to investage their is relationships between the variables

Correlations don't give information about the cause an effect-they only tell us if the relationship exits.

20 of 44

Pilot studies

A pilot study is a small version of the study in order to assess the effectiveness and makes changes to the study.

Resons why a pilot stduy is useful in research:

  • It ensures that the instructions are not too complicated for the participants so they understand and not to leave any importtant steps are taken out
  • It allows the researcher to identify flaws which gives the researcher a chance to re-design or improve parts of the study to avoid difficulties.
  • They save time and money as pilot studies provide the researcher enough data to decide of they want to go ahead with the study.
  • Allows the timescale of the experiement to be estimated.
21 of 44

Other research methods

Case Studies are a detiled study of a single individual. It uses information from a wide rnage of sources such as the person concerned and friends and family. The findings are orgnaised to represnet the individual's thoughts, emotions, experiences and abilities.

Case study provide rich data so the information may be overlooked by other methods.            They are useful in investigating human behaviour for example cases of people with brain damage.

Case studies may not be generaliseable because they may involve past events so they may be unreliable.

Content analysis is an anaylsis of a content of something. For exmaple the researcher may study gender content of magazine advertisements. Samplying and behavioural cetgories can be used.

Content analysis has high ecological validity as it's based on the observation of what people actually do.

Observer bias may occur because different observers may interpret the meaning of the behavirual categories differently.

22 of 44

Central tendency and dispersiion

Meaures of central tendency is about the values of the data. This can calculated in different ways.

Mean is adding up all the data items together and divding by the number of items used

Mean is likely to be representative of all the scoresIt's harder to calulate than all meausres-cannot be used for nominal data as it can be affected by extreme scores.

Median is all the items ranked from the smallest to the largest and then pick the middle value.

Median is unaffected by extreme scores and easier to calculate than mean                             It's less representative as it does not use all the scores.

Mode is the most common value

Very easy to calculate and unaffected by extreme scores

It's not very a useful way in describing data -it does not use all the scores.

23 of 44

Central tendency and dispersion

Dispersion describes how spread out the items are.

Range- is done by taking the lowest value from the higest value

Range is easy to calculate and no affected by extreme values.

It's not representative of all the data.

Standard divation- is the measure of the average between each value

It's representative as it takes all the scores into accont

Its harder to calulate than range

24 of 44

Problems with experiments and how to deal with the

Demand characteristics- The participants are aware of what the researcher is looking for. This can lead to demand characteristics where the participants may confound the results by trying to help the researcher which can mess up the data.

Single blind design-The researcher can prevent demand characteristics by not telling them the aim of the study before taking part.

Investigator effects-The presence of the researcher can have an influence on the results gathers because it changes the participant's behaviour.

Double blind design-The researcher can prevent investigator effects by using a double-blind procedure. By doing this the researcher and the participants are.

Randomization- Avoiding the effect of bias by designing the material and deciding the order of conditions.

25 of 44

The role of a peer review

A researcher investigation must go through a peer review process. A peer review is a written up academic paper by the researchers. Researchers suggest can improvement can be made to the investigation. As the editor will be receiving comments from a number of reviewers.

If research was published without this process of review and checking, poor research might be disseminated which would damage the integrity of that field of research.

Strengths of Peer Review

Peer review promotes and maintains high standards in research, which has implications for the society and funding allocation so it's assigned to high-quality research.

Helps to prevent scientific fraud, as submitted work is examined closely and thoroughly

Weaknesses of a Peer Review

If anonymity is not maintained experts with a conflict of interest might not approve research to further their own reputation or career.

26 of 44

Implications of Psychology researchon the economy

Applying topic in the book                                                                                  Psychologists have had a huge contrubution to the understanding of a number of issues.

Attachment-Bowbly has demenstrated the importance of attachment during childhood. Attachment has lead to the development of new working practices for parents such as encouraging them to have strong emotional bonds with their children. Its also changed the way how children are treated without their parents such as encourging early adoption and ensuring children get good phsyical care.

Improving memory- The use of cognitive interview has improved the accurate amount of information collect by eyewitnesses. Alsom cogntitve interview has been able to reduce expenses on wrongful arrests and ensuring criminals are caught.

Social change-Social change has been able to improve people's behaviour. For example, campaigns have been development to promote healthy eating which encourgaes positive behaviour change.

Psychopathology-Studies have compared the effectibess of different therapies. If research has show people with depression are likely to relaspe after having CBT, CBT may be more exspensive than drug therapy. People shoould be offer CBT because people take less time off work.

27 of 44

Demand Characteristics

The presence of demand characteristics in the study suggests that there is a high risk that participants will change their natural behaviour in line with their interpretation of the aims of the study and this affects how they respond to the task they have set.

Participants may, for example, try to please the researcher by doing what they have guessed is expected of them.

Repeated measures design- Demand characteristics will occur in a repeated measure design as the participants take part in the all the conditions, so they will be given enough information on the real purpose of the study. 

Independent groups- Independent groups design to lower the risk as participants only take part in one condition so they will be less likely to know the aims of the study.

Observational studies-Observational studies are also generally less likely to present demand characteristics, as participants might not know that they are taking part in a study.


28 of 44

Demand Characteristics

Hawthrone effect- The Hawthrone effect is when there is increased the attention given by the researchers. This can lead the participants to try harder than they usually do.

Social desirability-This is when the participants may present themselves in what they deem socially acceptable fashion.

For example, when asked if they have ever stolen anything, participants may lie in order to avoid presenting themselves in a bad light, consequently leading to gathering inaccurate data.

29 of 44


Reliability- Reliability is the measure of whether something stays the same i.e is consistent. The results of psychological investigations are reliable if they stay the same.

Reliability can be split into two main branches: internal and external reliability.

Internal validity- Internal validity describes the internal consistency of a measure such whether the questions are different in the questionnaire are measuring the same thing. 

External reliability- External reliability is assessing the consistency when different measures of the same things are compared. i.e. does one measure match up against other measures?

30 of 44

Improving reliability in observations

Observations: reliability can be improved by making sure behavioural categories have been properly operationalised and that they are measurable/self-evident.

Categories shouldn't overlap and all possible behaviours should be covered on the checklist.

The observers need more  practice using behavioural categories so they cannot respond more quickly

31 of 44

Improving reliability in Questionnares

Questionnaires: Reliability of questionnaires should be measured using the test-retest method.

The designer of the questionnaire gives the test to a group of people and gives the same people the same test at the second time. There will be a short interval between the tests such as a week or two because people will not remember their answers. The scores of each person are compared to see if there are similarities in the responses.

32 of 44

Improving reliability in Interviews

Interviews: Best way of ensuring reliability is making sure the same interviewer is used each time-inter-interview reliability.

A researcher could assess the reliability of one interviewer by comparing the answers on one occasion with answers from the same person with the same interviewer a week later. 

33 of 44

Improving reliability in experiments

Experiments: Lab experiments are described as reliable because the researcher has strict control over the procedure, for example, the instructions that the pp's received and conditions within which they are tested.

Standardisation: The procedure should be repeated for different participants and this is important as the procedures are exactly the same each time because otherwise, we can't compare the performance of the participants. For this reason, the performance must be standardised. 

34 of 44


Validity refers to whether a measure actually measures what it claims to be measuring

External validity is a measure of whether data can be generalised to other situations outside of the research environment they were originally gathered in

Internal validity is a measure of whether results obtained are solely affected by changes in the variable being manipulated.

35 of 44

Assessment and improving validity

Face Validity: Basic form of validity, whether a test, scale or measure appears to measure what it is meant to.

Simply looking at the measurement or passing it onto an expert to check.

Concurrent Validity: demonstrated when the results obtained from a test or scale are very close to or match those obtained on another recognised or well-established test. 

You compare your results to a well-known test to check for validity.

Close agreement between the tests would indicate the new test has high concurrent validity 

36 of 44

Improving validity in experiments

The use of single-blind and double-blind procedures is designed to achieve the same aim. 

Single-blind: Pp's unaware of the aims of the study which reduced the effect of demand characteristics on their behaviour.

Double-blind: the Third party constructs the investigation without knowing it's the main purpose which reduces demand characteristics and investigator effects. This then improves validity.

Using a control group can improve validity as the researcher is able to assess whether the changes in the DV were due to the effect of the IV.

37 of 44

Type 1 and Type 2 errors

A type 1 error: When the null hypothesis is rejected and the alternative hypothesis is accepted. It should be the other way round because, in reality, the null hypothesis is 'true'. It is also referred to as an optimistic error or false positive as the researcher claims to have found a significant difference or correlation when it doesn't exist.

A type 2 error: When the null hypothesis is accepted but it should have been the alternative hypothesis because, in reality, the alternative hypothesis is true. This is known as a pessimistic error or false negative.

More likely for type 1 error to occur if significant levels are lenient(too high)= 0.01 or 10%

Type 2 error more likely to occur if significant levels are stringent(too low)=0.01 or 1%

38 of 44

Types of data

Qualitative data: Qualitative data is non-numerical and includes descriptions in a verbal or written form.

Qualitative data provides detailed information which can provide insights and thoughts into behaviour because the answers are not restricted to previous expectations.

However, some data may oversimplify reality, for example in a questionnaire with closed questions it may force people to tick questions that don't represent their feelings. Therefore conclusions are meaningless 

Quantitate data: Quantitative data is numerical. This involves information that represents how much or long or how many i.e behaviour is measured in numbers

Quantitative data is easy to analyse because it uses descriptive statistics. This enables conclusions to be easily drawn. 

39 of 44

Types of data

Primary data: Primary data is information observed or collected directly from first hand.

The research is able to have control over the data. The data collection can be designed so it fits the aim and hypothesis of the study.

However, primary data is lengthy and expensive 

Secondary data: Secondary data is information in a research study that was collected by someone else or for the purpose of another current one.

Secondary data is cheaper because there less time and equipment needed.

However, the data may not fit the needs of the study

40 of 44

Psychology and the economy

Biopsychology: Neuroscience helps psychologists understand the human brain. the American government suggests that this may have practical economic benefits in the area of smart machines i.e machines that will think like humans. This does not mean building human alike robots but money saving intelligent machines for example questions on a telephone

41 of 44

The use of statistical tests

To find the appropriate critical value you need to know:

1. Significance level selected usually p <0.05

2. Kind of hypothesis; is the hypothesis one-tailed (directional) or two tailed (non-directional)

3. Value of N= Number of participants in the study

42 of 44

Non-parametric tests of difference

Wilcox tests: if the calculate is equal to or less than this critical value our results is significant 

Mann-Whitney test: if the calculated value if equal to or less than the critical value our result is significant.

Chi-squared test: if the critical value is equal or greater than this critical value the results is significant

The sign test: 

1. state the hypothesis: one tailed or two tailed

2. record the data of each sign add the + and add the -

3. find the critical value: by adding the + and - you select the smallest value  

4. Find the critical value of S: N= The total number of scores  

5: find the result of the direction: if the hypothesis is directional we can check the result is in the expected direction 

43 of 44

Non-parametric tests of difference (Sign test)

6. Report a conclusion: the calculate value of S must be equal or less than the critical value

44 of 44




I found these revision cards really helpful !!

Similar Psychology resources:

See all Psychology resources »See all Research methods and techniques resources »