Psychology - Experiments, Ethics, Sampling, Aims and Hypotheses

RESEARCH METHODS

a range of ways to carry out scientific investigations

RELIABILITY

This refers to how consistent a study or measuring device is. A measurement is said to be reliable or consistent if it can produce similar results if used again in similar circumstances. As a psychologist you have to ask yourself ‘Did every participant experience my study in the same way?’ If they can say yes to that question, they have high reliability.

VALIDITY

As a psychologist you have to ask yourself ‘Is my study testing what I want it to test?’ Having high validity can also be thought of as collecting data that truly reflects how a participant would behave. A psychologist must ensure that their participants aren’t effected by anything else, other than what they are testing. They may ask ‘Are my participants lying? Are they changing their behaviour due to my presence?’ If the answer is no to these questions, they have high validity

External Validity

Whether the findings will generalise to other populations, locations, contexts and times and still hold true.

External Validity: Ecological validity AKA Mundane realism

Refers to the extent to which the findings of a research study are able to be generalised to real-life settings.

External Validity: Population Validity

How representative the sample used is to other populations.

External Validity: Historical/Temporal Validity

Will the findings still be valid as society changes over the years e.g. will a study conducted about female behaviour in 1965, generalise to females today?

Internal Validity

Within your measure, the IV is the only variable effecting the DV.

Internal Validity: Face Validity

The degree to which a procedure, especially a psychological test or assessment, appears effective in terms of its stated aims.

Internal Validity: Concurrent Validity

Whether a measure produces similar results for a participant as another test that claims to measure the same thing e.g. a participant completes a brand-new test for autism and gained very similar results in a previously well-established test of autism.

Experiments

Experiments are the only research method that manipulates variables. We manipulate one variable (IV) and see if it could have an effect on another variable (DV) (cause and effect)

Experiment using a Self-Report

Experiments can be used alongside other research methods; . For example a self-report can be used as your dependent variable. A self-report is when you ask people how they think and behave. This can be done through interviews questionnaires. Psychologists then make conclusions about how people think and behave from their answers.

example of experiment using a self report:

I want to know whether the temperature of a room has an effect on people’s ability to concentrate.

For example: Ask people to rate from 1-10 how hard they find it to concentrate (using a questionnaire as the dependent variable) in a warm or a cold room (this would the independent variable, as it is the thing that changes).

Experiment using an Observation

Experiments can also use an observation as the dependent variable. An observation is where you watch how people behave. Then psychologists make conclusions based only on what you can observe

example of experiment using an observation:

I want to know whether the temperature of a room affects people’s ability to concentrate.

For example: Go to a library and observe how many books they read/how distracted they get/how long they stay for, when the temperature is cold compared to warm

ethics:

a moral code that psychologists follow in order to protect both the participants and researchers.

explanation of ethics:

Psychologists are obliged to consider the psychological well-being, health, values and dignity of their participants. If they do not do this properly, their research is described as unethical. This would put psychology is an unfavourable light and would hinder the ability to replicate the research. Researchers should strive to ensure that their research is as ethical as possible.

how is ethics guided?

Guidelines are issued by the British Psychological Society (BPS). These clarify what is ethically acceptable in psychological research.

mnemonic for ethics:

Can Do, Can’t Do With Participants In Psychology

CAN:

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

DO:

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.

CANT:

Competence – having the knowledge and skills, and attitudes, values, and judgment needed to perform the work of a psychologist

DO:

Debrief – Full explanation of the aims and potential consequences of a study are given to the participants straight after the study has finished. The debrief is extremely important because it fulfills many other ethics in the process. For example, in the debrief will advise participants that their data is to be kept confidential. They will also be given another opportunity to withdraw their data after the event even if they consented to the survey and were aware of the aim at the start. The psychologist will ensure that no one has been harmed and if so, will be offered additional support.

WITH:

Withdrawal – Participants should be made aware that they can leave at anytime and can remove their data from the study at anytime.

PARTICIPANTS:

Protection from psychological and physical harm – avoid harming participants mentally/physically or psychologically – Participants should leave in the same positive state when they finish the study, as when they arrived.

IN:

Informed consent – Having sufficient knowledge about the study to be able to make an informed decision to participate. After being informed about the study, participants must agree to take part before the study.

PSYCHOLOGY:

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 someones window, without their knowledge.

Target Population

The set of people researchers want to find out about

Sample

A small set of people taken from the target population

Representative

How well a sample reflects the target population

Sampling Bias

Samples can be biased; they do not reflect the target population and this affects the conclusions we can draw from these samples.

Androcentric

a sample that contains a large proportion of males.

Gynocentric

a sample that contains a large proportion of females.

Cultural bias

a sample that is too focused on one culture, isn’t representative of all cultures

Ethnocentric

This is when research is generalised (trying to apply) to other cultures without considering how cultures are different

Population validity

Being able to generalise results from our sample to the target population and still hold true.

Sampling Methods/Techniques:

1. opportunity sampling

2. volunteer/self selected sampling

3. random sampling

4. stratified sampling

5. systematic sampling

Opportunity Sampling:

Anyone who is available at the time of your research e.g. I want to research into the eating habits of men. I walk around the college and survey the first 20 males I find

Strengths:

1. Quick and easy to carry out. This is because it relies on people who are around at the time and is therefore more time efficient than other sampling methods.

2. Can help to collect participants with similar characteristics as people who share characteristics tend to segregate in the same areas. This is a strength because it can help to generalise (apply) findings to a target population.

Weaknesses:

1. May not always be representative, as the kinds of people available are likely to be limited, and therefore similar, so although they may apply to the target population, it may be difficult to generalise to the wider population.

2. Increased chance of researcher bias as they may only approach people who they feel will give them the results they want.

Volunteer/Self-selected Sampling:

Participants choose themselves to take part in the study. They could be recruited through; using online email surveys, signing up or applying to take part, or responding to adverts or posters.

Strengths:

1. This method is relatively easy as the participants come to you. This is a strength because they are likely to remain committed to the study and less likely to drop out, preventing the chance of a small, unrepresentative sample.

2. Can reach a wider variety of participants through emails, posters, advertisements compared to opportunity sample, which will only cover a small area.

Weaknesses:

1. This method may have an issue with sample bias. This is because certain individuals tend to volunteer for studies and therefore may not be representative of all people. For example if I wanted to investigate the effect of exercise on mood, people who are good at exercise are likely to come forward. Therefore, the results would not be valid when generalising to people who do not exercise.

2. Sometimes there will not be enough interest in your studies advert which can lead to a small sample. This is a weakness as it could lead to an unrepresentative sample.

Random sampling:

Every member of the population has a fair and equal chance of taking part. e.g. Everybody puts their name into a hat, draw first 25 names out of the hat for the sample.

Strengths:

1. This is the most representative sampling technique to use as all types of people in the population have an equal chance of being chosen. If you have a wide variety of differences in your sample, it is more likely to be generalisable to the wider population (i.e. other people outside of the target population)

2. It can provide an unbiased sample as the researcher has no part in deciding who is selected, therefore reduces the chance of researcher bias, increasing validity.

Weaknesses:

1. Time consuming and hard to ensure that everyone is equally chosen e.g. due to lack of information or access.

2. Sample could still be biased e.g. if only girls happen to be selected, this would create a gynocentric sample that lacks generalisability.

3. It can be impractical (or not possible) to use a completely random technique, e.g. the target group may be too large to assign numbers to. Therefore there tends to be some choice from the researcher in terms of narrowing the sample down.

Stratified Sampling:

Here the sampler divides or ‘stratifies’ the target group into sections, each showing a key characteristic which should be present in the final sample. Then each of those sections is sampled individually. All types of members of the population are represented by deliberately selecting participants from all strata e.g. elderly, middle-age, young adults, teenagers, children. The sample thus created should contain members from each key characteristic in a proportion representative of the target population. For example, if there were strata’s of the class system – upper, middle and lower class, there would need to be a much larger sample for the lower class sample compared to the upper class, as there are a lot more people in the lower class population compared to upper class.

Strengths:

1. This sample is likely to be more generalisable and this is because it tries to gain a wide variety of people. The more diverse a sample is, the more generalisable it is to the wider population.

2. Assuming the list order has been randomised, this method offers an unbiased chance of gaining a representative sample

Weakness:

1. Difficult, as all the subgroups in the population must be known and accessible, this is often difficult to achieve.

2. It takes more time and resources to plan.

Systematic sampling:

A systematic method is chosen for selecting from a target group, e.g. every fourth person in a list could be used in the sample. It differs from random sampling in that it does not give an equal chance of selection to each individual in the target group.

Strengths:

1. It can provide an unbiased sample as the researcher has no part in deciding who is selected, therefore reduces the chance of researcher bias, increasing validity.

2. Higher chance of gaining a representative sample as it is likely to gain a varied sample from the systematic formula. It would be quite unlucky to only pick females for example.

Weaknesses:

1. If the list has been assembled in any other way it could unintentionally create a biased sample. For example if every fourth person in the list was male, you would have only males in your sample.

Experimental Designs:

how to group your participants in an experiment

Independent measures design:

This is when different participants participate in each condition (each level of the IV). For example, having Dave, Steve and Julie complete a condition where they eat chocolate and then rate their happiness on a scale of 1-10, then a different group of people; Harry, Jerry and Phil complete a condition where they have no chocolate, and then rate their happiness on a scale of 1-10.

Strengths:

1. Different participants are used in each condition, so there are no practice effects. For example if I wanted to see whether an energy drink would have an effect an individuals ability to shoot some hoops compared to not having an energy drink. It may not be energy drink that effected their ability, but the fact that they’ve had two attempts at shooting hoops. Independent measures design avoids this.

2. Participants only see the experimental task once, meaning that they are less likely to guess the aims of the study and change their behaviour as a result. For example the participant may work out that the researchers are trying to see if energy drinks increase sport performance and then change their behaviour to fit in with their expectations. This is known and demand characteristics and would reduce the validity of the findings.  Demand characteristics can be further reduced by carrying out a single blind or a double blind procedure.

Single blind test 

This is when the participants are unaware of the condition that they are in. This means they are less likely to guess the aim of the study as they have not been given reasons or explanations as to the condition they are in. Therefore reducing demand characteristics.

Double blind test 

This is when neither the researcher or the participants are aware of which condition an individual is in. This ensures that demand characteristics are reduced from the participants, but also researcher bias is also reduced. Researcher bias is when the researcher intentionally or unintentionally influences the behaviour of the participants in order to get the participants to fit in with the results that they want. If the researcher is unaware of what condition it is, they would find it more difficult to influence.

Weaknesses:

1. Individual differences could act as an extraneous variable (an extra variable that you do not want to have an effect on your DV. This is because you only want to measure the effect of the IV in order to establish cause and effect) and make it look like the IV effected the DV, but in fact it was something about the individual. For example, if everyone in the condition of 5 people present were on the basket ball team, and the group with 1 person present had participants that didn’t play sport, it could look like crowds of people improve sport performance, but in reality, it was due to the participants previous shooting hoop skills.

2. More participants are needed, so maybe more difficult to find.

extraneous variables

These are variables that could potentially effect the DV when we don’t want them to. We as researchers only want the IV that we have created to effect the DV. There are different types of extraneous variables:

Situational variables

things that could effect participants behaviour in the environment, such as lighting, sound, temperature.

Design variables

things like order effects, fatigue effects i.e. tiredness by carrying out a repeated measures design. For a independent measures design, you have the issue of comparing two different groups of people, so there will be participant variables.

Participant variables

things like mood, intelligence. Things about the participant that may effect their behaviour in the study.

Repeated measures design

When the same participants, participate in each condition.

Strengths of Repeated measures design

1. Individual differences / participant variables will not distort the effect of the IV on the DV, as participants do both levels. For example if I wanted to see whether classical music compared to rock music had an effect on math’s ability, I would not have to worry about the participant variable of math’s ability. This is because I would be comparing results from the same person. Unlike in an independent measures design where I could be unintentionally comparing participants with really bad math’s ability to those with really good math’s ability.

2. If counterbalancing is used, this can reduce order effects as outlined in the evaluation of independent measures design.

counterbalancing

Counterbalancing works by splitting a condition of participants in half and making them experience all possible orders of the study.

How does counterbalancing resolve order effects

The participants who did the conditions in the close then distant order have experienced the order effects of that specific order. The participants who did the conditions in the distant then close order have experienced the order effects of that specific order. Counterbalancing works when both groups have been effected by the order effects equally as much as each other. This means that the order effects are cancelled out

Weaknesses of a Repeated measures design:

1. Order effects such as practice (like the example above), and fatigue effects i.e. participants get bored of doing the same thing and so do not contribute a valid reflection of their behaviour.

2. Participants see the experimental task more than once meaning they are more likely to guess the aim, and therefore more likely to suffer from demand characteristics.

Matched pairs design:

This is when different participants participate in each condition. However, each participant in one group is matched on a certain characteristic to another participant in the other group. The matching is done on relevant variables. For example: an experiment investigating the effect of sleep on Maths ability. Participants maybe matched on their Maths GCSE grade to ensure that it doesn’t act as a participant extraneous variable.

Strengths of a Matched pairs design:

1. Different participants are used in each condition, so there are no order effects

2. Participants only see the experimental task once, meaning that they are less likely to guess the aims of the study, and therefore reduce demand characteristics.

3. The effects of individual differences are highly controlled, so less chance of participant extraneous variables.

Weaknesses of Matched pairs design:

1. The similarity between the matched participants in each condition may be limited. This is because it is difficult to ensure that the matching is completely accurate, as other variables about the participant may be unknown. For example, you may match someone on Maths ability for students who got a B in Maths GCSE, but one of the participants works in a job where they use Maths everyday, but the other has a job that does not require Maths on a daily basis and hasn’t used certain types of Maths since school.

2. Matching participants can be very time consuming and difficult.

Experiments

This is a research method that always uses an Independent Variable (a variable that you change) and a Dependent Variable (a variable that you measure).

Strengths of experiments:

1. Using a repeated measures or matched pairs helps to reduce participant variables

2. Using independent measures design helps to avoid the influence of demand characteristics and order effects.

Weaknesses of experiments:

1. If participants are unaware that they are in the study, this raises ethical issues.

2. The researchers expectations and interest in their research topic may lead them to collect biased results i.e. only collecting results that would fit in with what they want to find.

laboratory experiment

This is a study that is carried out in unnatural settings with a highly controlled environment. This means that researchers can control for things such as the extraneous variables mentioned earlier. Lab experiments are also highly standardised. This means that participants experience the same instructions, environment, test, researcher etc. This links in to reliability, being highly standardised will mean that you measure is consistent. By having high control and high standardisation, researchers can be more confident that the IV EFFECTED their DV. It also means that the study is easy to replicate, as the instructions on what happened within the study is very clear and standardised.

Strengths:

1. There is high control of extraneous variables – therefore increasing confidence that the IV EFFECTED the DV. This is known as Cause and Effect.

2. Standardised procedures and instructions are used which enables researchers to repeat the study in the exact same way with other participants. If the results are the same with different people. We can be more confident that the IV effected the DV.

3. In a repeated measures design, counterbalancing can be used to reduce design extraneous variables such as order effects.

Weaknesses:

1. The unnatural/artificial situation of a lab may mean that participants do not display behaviour that reflects how they would behave in real life – this is known as low ecological validity.

2. Due to having a researcher present, participants may suffer with demand characteristics. This means that they behave in a way that would please the experiment i.e. fit in with what they are trying to find out. This again means that the participants are not behaving in true way.

Field experiments:

In a field experiment, there is still an IV and a DV. but the setting of the experiment is normal in relation to the behaviour observed, for example, carrying out a study on shopping behaviours in a local super market.

Strengths:

1. As participants are in their normal environment/situation, their behaviour is likely to be more valid, as it will reflect their true behaviour.

2. Participants may be unaware that they are being studied, and therefore less effected by demand characteristics.

Weaknesses:

1. Control over extraneous variables is more difficult because the situational extraneous variables are difficult to control. This can make them less reliable and difficult to replicate in a standardised way.

2. The researcher cannot be sure that the IV caused the effect on DV. This is due to the lack of control over the environment.

3. Participants could be unaware that they are being studied as many field experiments are in natural environments such as supermarkets, schools etc. This raises ethical issues.

Quasi experiments:

This is when the researcher does not manipulate the IV because it is naturally occurring within the participant. For example age, gender, disability. A researcher cannot change your age, disability etc, therefore the IV is created by creating groups of participants with different levels of the IV. For example I might want to find out if age effects concentration, as a researcher I can’t change someones age, therefore, I use a quasi experiment and create the following groups; (Young – 15-25, Old – 26-50). Researchers also use this method when it would be unethical to manipulate the IV. For example when comparing witnesses to real crimes who have been ‘frightened’ and ‘not frightened’, it would be quite unethical for me to stage a serious crime in front of them and then ask them about it whilst being deceived. Therefore, I would search for people who have witnessed a crime in their past and have claimed to be either ‘frightened’ or ‘not frightened’ and use them in separate groups.

Strengths:

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

2. They allow researchers to investigate variables that would be unethical to manipulate.

Weaknesses:

1. 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.

2. They can be difficult to achieve large sample sizes, particularly if the IV is something uncommon such as a rare mental health issue.

3. Quasi-experiments can only use independent measures design and therefore there is an increase in the chance of participant extraneous variables.

Natural Experiments:

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

For example, a psychologist noticed that a new leader in Madagascar had a huge impact on poverty. They decided to carry out a natural experiment by comparing crime rates (DV) before the leader was elected (decreased poverty) and after the leader was elected (increased poverty)

Strengths:

• Behaviour 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.

Weaknesses:

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

• Due to the lack of control of extraneous variables it is more difficult to establish cause and effect.

Research aim:

what you aim to find out. For example: A study investigating the effects of chewing gum on memory recall

Research question:

pretty much the aim, but phrased as a question. For example: Does chewing gum effect memory recall
Hypothesis: Specific, testable prediction of how one variable affects another.

different types of hypotheses?

• Directional hypothesis

• Non-directional hypothesis

• Null hypothesis

Non-directional hypothesis

In any non-directional hypothesis, you must use the word EFFECT. This is because we predicting that there will be an effect, but we are not predicting the direction of the effect, as it could go either direction.

Directional hypothesis

In a directional hypothesis we tend to use words, such as; “POSITIVE, NEGATIVE, FEWER, HIGHER”. This is because we are predicting which way the effect is going.

Null hypothesis

predicts no EFFECT. It is null/void/zero and therefore we tend to use the word ‘NO’ in a null hypothesis. When writing hypotheses, psychologists will always write a null hypothesis and a directional or non-directional hypothesis. This is because at the end of study when they analyse their results, they must decide which hypothesis to accept or reject .i.e. did an effect happen, or did it not?