not this one allie Flashcards

Question

Methods of investigation **24.** **Field Experiments** **Blue card**

Answer 1

In a field experiment the psychologists manipulates the independent variable (IV)but the experiment takes place in a real life setting for example, a university, school, summer camp (E.g. Sherif et al Robbers Cave; Eliot’s Blue eyes brown eyes; Piliavin’s study into ‘Good Samaritans)

Answer 2

Disadvantages of field experiments 1. Deception: because participants do not know that they are taking part in an investigation in the majority of field experiments, the researchers are deceiving them. However deception is often necessary. 2. As the participants are being deceived, the researchers do not have their consent, let alone, full informed consent. However, this cannot always be obtained, as demand characteristics would then be present, and real behaviour would not be demonstrated.

Answer 3

Advantages of field experiments 1. Because the study is taking place in a natural setting, the behaviours is more likely to be more natural compared with a laboratory experiment; there is a higher ecological validity 2. There is less chance of demand characteristics – with most field experiments the participants do not know that they are taking part in an investigation.

Answer 4

The quasi or natural experiment is one in which the independent variable occurs in real life, so the researcher cannot ‘create’ a difference for the purpose of the experiment. In a quasi experiment the IV may be age, gender or race. In a natural experiment the IV may already be occurring. ----------EG the effect of the media on eating disorders. Used a Caribbean island where TV had not been introduced and where there were no diagnoses of eating disorders. However a few years later when TV was introduced a number of girls had been diagnosed (Becker et al.) This is an example of a natural experiment, as the behaviour occurred naturally. The researchers did not manipulate anything. Some examples of what could be investigated using a natural experiment are: How people cope in a hurricane / The effect of cigarette sales after the ban on smoking ban (in public places) in the UK [consider how you might conduct such an experiment]

Answer 5

Advantages of natural experiments: 1. Natural experiments allow researchers to investigate behaviours that, for ethical reasons, could not be studied or created. 2. Because participants are unaware that they are taking part in an investigation, there will be less chance of demand characteristics and more natural behaviour. Disadvantages of natural experiments: 1. Because the independent variable is not controlled by the investigator, there is less control from the researcher’s point of view.

Answer 6

An experiment is usually organised so there are two trials, after which the performances of the participants are compared. These are the **conditions** of the experiment.

Answer 7

The way in which participants are assigned to groups is called the experimental design. How they are used. There are usually 2 conditions in an experiment and the researcher has to decide who will take part in each condition. (E.g. researchers might want to know if one group of people are better at something than another group so might have 2 groups. Sometimes though 1 group experiences the independent variable and the other group (the control group) does not. )

Answer 8

The way in which participants are assigned to groups.

Answer 9

An experiment is usually organised so there are two trials, after which the performances of the participants are compared. These are the conditions of the experiment.

Answer 10

1. Independent groups design 2. Repeated Measures design 3. Matched-pairs design.

Answer 11

The available people for the experiment are divided into 2 groups. Each group is independent of the other. 1 group takes part in one of the conditions of the experiment; and the other takes part in the remaining condition of the experiment. E.g. we could test the effect of different music (the IV) on a person’s recall by splitting the experimental group in two. Grp 1 could listen to pop music and group 2 to classical music. Both groups would be given the same memory test. The control group would have the memory test but no music would be played. In laboratory and some field experiments the researcher is able to choose which participants are assigned to the experimental group and which to the control group. Assigning them to groups should be by random allocation so that everyone has an equal chance.

Answer 12

It’s the quickest and easiest way of allocating participants; Because they are assigned randomly to groups there are no order effects (which occur with repeated measures) Often the same material can be used for the tasks in both conditions Participants cannot work out the aim of the study because they only take part in one condition.

Answer 13

Having different people in each condition could be why the results are different: People vary in their experiences, attitudes, intelligence, alertness, moods and so on. These are participant variables. Because participants are assigned randomly the researchers do not know, for example whether one group comprises most of the more alert or skilled participants. If this were the case these participant variables might produce differences between the results from the two groups that are not due to the independent variable. This is why a larger sample is needed to reduce this effect. You need more people for the study.

Answer 14

Randomly assigning participants into different groups

Answer 15

**Participant variables** are the differences between the people who take part in the study. This may affect the results of an experiment that used as independent group design. E.g. people vary in their experiences, attitudes, intelligence, alertness, moods and so on. These minor differences can cause a **random error** in the results. An error that occurs by chance.

Answer 16

There is 1 group of participants. Every participant goes through both experiences (the experimental condition and the control condition). For example the participants taking part in the music and memory test would experience all 3 conditions, pop music, classical music and no music. To combat order effects or demand characteristics on the results counterbalancing should be carried out.

Answer 17

Because the same people are in both conditions, there are no participant variables. You only need 10 people to get 20 results because each person produces two ‘scores’ requires fewer participants, as the participants are in both conditions.

Answer 18

There are drawbacks because it can produce biased results. The participant may guess the aim of the study when they take part in the second condition, so demand characteristics might affect the results. Taking part in both conditions may produce order effects, or practice effects, and may affect participant’s performance.

Answer 19

Experiencing both the experimental and control conditions in a repeated-measures design may affect participants’ performance by damaging it (because they have become bored or tired of repeating the task) -- Order effects: or improving it (because they have already done a similar task -- the practice effect.

Answer 20

To combat the influence of order effects or demand characteristics on the results, counterbalancing should be carried out: this means reversing the order of the conditions for the participants. The sample is split in 2: one half does the experimental condition (A) and then the control condition (B); the other half does the control condition (B) and then the experimental condition (A) ABBA design. Although order effects will still occur- they will occur for both groups and balance each other out. In the music example we could have half of the group: no music; then pop/classical; the other half pop/ classical and then no music condition.

Answer 21

Counterbalancing means reversing the order of the conditions for the participants (to combat order effects or demand characteristics).

Answer 22

Available people are matched for qualities into pairs on the basis of variables relevant to the study, such as age, gender, intelligence, reading ability or socioeconomic background. This may require pre-tests in order to ensure good matching, and then one of each pair is randomly assigned to the experimental condition and the other to the control condition. The perfect matched-pairs design is one that used identical twins, assigning one to each condition.

Answer 23

**Strengths:** Because participants experience only one condition of the experiment there are no order effects (so no need for counterbalancing) Often the same material can be used for the task in both conditions **Weaknesses:** This method can be more expensive and time consuming that the others and is not always successful Some participant variables are still present

Answer 24

Now you need to consider where you will get the people from to take part in the experiment The sample is made up of the participants in the research. They should be as REPRESENTATIVE as possible of the target population. The more representative the sample the more confident the researcher can be that the results can be GENERALISED to the target population. However in reality very few samples are truly representative because of the difficulties of sampling methods (choosing participants). The sampling methods we could use are: [ROSS] - Random sampling; - Opportunity Sampling - Systematic sampling - Stratified sampling.

Answer 25

The large group of people the researcher wants to study. Methods of control, data analysis and date presentation

Answer 26

The sample is the participants used in the research. Sampling methods are the ways of choosing participants. This includes Random sampling Systematic Sampling Stratified Sampling

Answer 27

In a random sample *every member of the target population* has an equal chance of being selected for the sample. This means that the researcher must identify all the members of the target population, number each person and then draw out the required number of people. If the target population is small, then all the numbers can go into a hat to be drawn out. If it is large, then a computer programme can be used to do this. This sampling method is fair and not biased because the researcher cannot choose the individual participants. However this sample might still not be representative because, for example, the researcher could draw out too many females, just by chance. E.g. in a study with a target population of seven year olds, the names of every seven yr old from all the local primary schools are gathered, their names put on a slip of paper and drawn out of the box. Or they might all be given a number and a computer might select random numbers.

Answer 28

Every member of the target group has an equal chance of being selected. All members of the target population are identified and then selected out of a hat/ with a computer programme.

Answer 29

**Strength:** No researcher bias- Random sampling is the best technique for providing an unbiased, representative sample of a target population. **Weaknesses**: It can be very time consuming and is often impossible to carry out, particularly when you have a very large target population. How could you gather a random sample of people living in your area? How could you obtain the information, and how would you try to make the sample random? Methods of control, data analysis and date presentation

Answer 30

Researchers rarely use random sampling: they have to rely on participants who are more easily accessible. **Opportunity sampling means choosing people who are members of the target population and are available and willing to take part.** Often friends of the researcher so they may not represent the target population very well. This could produce a biased sample. They might try to ‘help’ the researcher so their results could be unreliable. However it’s a quick and easy way to collect people. Opportunity sampling occurs in Field experiments. E.g. we could have a confederate fall over at a football match to see how fans react. The sample of people would be football fans.

Answer 31

Opportunity sampling means choosing people who are members of the target population and are available and willing to take part (e.g. friends and in field studies). Methods of control, data analysis and date presentation

Answer 32

**Strength:** Quick, cheap and easy compared to other methods **Weakness:** As it only uses people who are available at the time it’s not likely to be representative and may have researcher bias

Answer 33

Systematic sampling means selecting participants at fixed **intervals from the target population**. Could be every third person from a class register or every 7th that comes out of a shop. Involves selecting every “nth” member of the target population. EG if the researcher decides that ‘n’ will be ‘7’ – every 7th person in the target population is selected. This is unbiased because the researcher cannot choose the individuals.

Answer 34

Systematic sampling means selecting participants at fixed intervals from the target population. ( “nth” member)

Answer 35

When carrying out systematic sampling its important that the potential participants are as randomised as possible. If you were choosing participants from a list of pupils in school arranged by class (yr 1, 2, 3,) you might end up picking out all the first years. This problem is called periodicity. It can be overcome in this example by mixing up all the names first.

Answer 36

**Strength/ advantage**: Fairly cheap and fast. Simple procedure with no researcher bias. **Weakness/ limitation**: Sample may not be representative. Researchers need to ensure that the list does not contain a hidden order (periodicity)

Answer 37

Provides a sample that is in proportion, in the relevant characteristic, to the target population. It’s the most complex of the sampling methods. The researcher must identify the subgroups in the target population and work out what proportion of that target population each group represents. E.g. conducting a survey of women to ask questions about their attitude to childcare. You pick out the relevant characterises as the types of work women do, so you need to find out what proportion of females in the whole population are: self-employed, professions, manual workers. Unemployed etc. (You can get this info from census data). These categories must be represented in the same proportion in your sample. E.g. If 10% are self employed then in a sample of 50, 5 should be self employed and so on. To get the required number in each category, participants should be random-selected from the target population.

Answer 38

To obtain this type of sampling, the different sub groups in the target population are identified; then people are randomly selected from these subgroups in proportion to their numbers in the target population.

Answer 39

**Strengths:** Provides a very representative sample **Weaknesses**: Very time consuming and expensive.

Answer 40

To be confident that the independent variable caused the dependent variable, the researcher must try to control all other aspects (or variables) in the experiment. There are ways that the research can increase control over the research environment and so have greater confidence that the IV has caused the DV.

Answer 41

In order to control variables, all participants must undergo the same experience, except for the independent variable. To ensure that all participants have the same experience, researchers should ensure that they are all tested: In the same place; with the same equipment and materials placed in the same way. Under the same conditions, e.g. level of lighting, noise and heat. At roughly the same time of day.

Answer 42

An important issue is how participants in a study are put into the conditions of the experiment. When the design is independent groups the experimenter can use random allocation to achieve this. E.g. each participant takes a piece of paper from a bag that determines the condition they are in. When the design is repeated measures (where participants take part in both conditions) this can cause order effects. In order to even these out counterbalancing is often used. Half of the participants complete condition 1 and then 2; the other half complete condition 2 and then 1. This shares the effects between the conditions,

Answer 43

If variables are not controlled, they may influence the results and so make them invalid. E.g. memory test for recall of words – if participants in 1 group saw the words but the participants in another group heard them read, it could account for the difference in the recall, rather than e.g. the way the words were organised. This is called a confounding variable. Another eg - traffic study-length of time to travel home bus or cycle – a confounding variable could be the fitness level of the cyclists.

Answer 44

A confounding variable is an extraneous variable whose presence affects the variables being studied so that the results you get do not reflect the actual relationship between the variables under investigation. Not controlling procedures leads to the possibility of extraneous variables occurring and confounding the results.

Answer 45

A variable that is not the IV but might affect the DV if it is not controlled E.g. in the music study experiment then EVs are things other than the presence or absence of music (IV) that might affect the learning score (DV). E.g. time allowed to do the test. If participants in the music present condition have more time than those in the music absent condition then that could affect the scores in the test. The solution would be to allocate the same amount of time for doing the test to both conditions.

Answer 46

The information that is said or written for participants might affect the way they do the study and therefore their scores or performances. Participants should be given identical instructions in exactly the same way. This is particularly important if the research requires them to perform a task. Sometimes instructions are written down, to safeguard against bias. But the problem is that some participants might have difficulty reading or understanding. Instructions need to be clear, simple and unambiguous.

Answer 47

Standardising instructions in the following ways: Briefing Standardised instructions Debriefing

Answer 48

Briefing This is what is said to encourage a person to agree to participate. It contains ethical information about consent, anonymity, the right to withdraw and so on.

Answer 49

Standardised instructions are clear instructions about exactly what the participant will have to do in the experiment.

Answer 50

Debriefing This explains the study in detail so that each participant is absolutely sure of the aim of the study, why they were doing what they were doing in the condition(s). Ethical issues are addressed again, especially the opportunity for the participant to withdraw their data if they feel unhappy about their performance.

Answer 51

Another technique used in practical work is called randomisation. This means making sure that, where necessary, there are no biases in procedures. E.g. in an experiment where a list of 20 words has been constructed for participants to learn. These words are of equal difficulty because they are all everyday nouns with six letters. The experiment has to decide the order in which they should be presented to the participants. Instead of the experiment choosing the order, randomisation is used. All 20 words are written on separate pieces of paper and put in a bag. The first word is pulled out and written down at the top of the list. This is repeated with all 20 words. The order of the words has now been randomised. Each word has an equal chance of being selected first for the list and experimenter left the final order of the words to chance.

Answer 52

Once the investigation has been conducted we need to look at the data that have been collected. For quantative data (numerical data) there are various calculations that can be used to analyse and provide some meaning at the end of a study. These include: mean median, mode range percentages.

Answer 53

The mean is often referred to as the average. Add up all of the scores in a condition and divide by the number of participants in the condition. E.g. in a study of 20 people standing on one leg for as long they could – you would add up all of their standing on one leg times and divide by 20 to get the mean standing on one leg time. The mean is useful because it makes use of all the scores. Sometimes though you can get a score that’s different to the others (an anomaly) and extreme. E.g. one person gets 0. This is known as anomalous. This can really mess up the mean.

Answer 54

The median is calculated by looking at the middle score in a set of data. First you have to place all of the scores in the order of value. In an odd number of figures the mean is the middle number. In an even number of figures the mean is the mean of the middle two numbers (i.e. take the two middle numbers add them together and divide by 2).

Answer 55

Mode is French for fashion. The mode is the most fashionable number. The most frequently score that occurs.

Answer 56

The range is a measure of dispersion. This means the difference between the highest and the lowest score in a condition.

Answer 57

If we want to work out the percentage for each participant we would divide the score they achieved by the total maximum score and then multiply by 100. E.g. for 45 out of 60 – we divide 45 by 60 x 100 = 75%

Answer 58

Once the data has been calculated, it is often useful to present it as a graph to make it easier for people to interpret and understand. A **bar chart** is one way to display data in categories. To make a bar chart. 1. Plot the categories along the horizontal (X) axis 2. Plot the frequencies (how often something occurs) along the vertical (Y) axis. 3. Don’t let the bars or rectangles touch each other! 4. Arrange them in ascending / descending order. 5. Label both axes. 6. Give an appropriate title

Answer 59

Once the data has been calculated, it is often useful to present it as a graph to make it easier for people to interpret and understand. **A line graph is often used to display data that are connected to each other to show the relationships between each point.** 1. List the categories along the horizontal axis (X) 2. List the frequencies (how often something occurs) along the vertical (Y) axis. 3. Plot the scores using a dot according to the frequencies 4. Join the dots together 5. Arrange them in ascending / descending order. 6. Label both axes. 7. Give an appropriate title

Answer 60

The experiment had to be stopped after 6 days because the ‘guards’ became sadistic and the prisoners became withdrawn and distressed.

Answer 61

Ethical issues are points of concern about what is morally right – desirable standards of behaviour we use towards others. If we behave ethically, then we treat others with respect and have concern for their well-being. Psychologists have legal and moral responsibilities to those who help them in their research: every individual has rights and these must be respected and protected. Participants put their trust in researchers. If their trust is betrayed this discredits the profession. The BPS has published guidelines specifying ethical concerns and how they must be addressed.

Answer 62

Ethical issues are points of concern about what is morally right – desirable standards of behaviour we use towards others. Psychologists have legal and moral responsibilities to those who help them in their research: every individual has rights and these must be respected and protected. The BPS has published guidelines specifying ethical concerns and how they must be addressed.

Answer 63

The Code of Ethics and Conduct of the British Psychological Society (2006) underpin the activities of all practising psychologists, researchers and psychology students. Therefore you, as a student, need to follow the BPS guidelines in everything you do. www.bps.org.uk

Answer 64

Psychologists should only give advice if they are qualified to do so. Gcse students are not.

Answer 65

Ethical considerations: Psychologists should respect people as individuals and avoid unfair or prejudiced practices.

Answer 66

The privacy of participants must be protected. This includes confidentiality, observations and the right to withdraw if participants feel uncomfortable. If people are being observed in a public place their privacy should be respected.

Answer 67

Participants, and the data gained from them, must be kept anonymous unless they give their full consent. It must not be possible to identity participants from any reporting of the research, such as in an academic articles or a newspaper article.

Answer 68

If participants are observed in public, in circumstances where anyone could observe them, privacy is not an issue. Having said this, if the observer is obvious to those being watched, perhaps taking notes, this may cause discomfort or distress, which is not acceptable. However when observers are hidden, then privacy is violated and this contravenes the BPS guidelines.

Answer 69

When participants agree to take part in research they do not know the extent to which you will encroach upon their feelings, emotions or sense of what is appropriate. Therefore researchers must gain informed consent and also tell participants that they can withdraw at any time during the research if they wish to do so. Participants should also be reminded of their right to withdraw if it is a long study or if they appear to be distressed. By reminding them of the right to withdraw you are stressing that they are under no obligation to continue and can act to protect themselves at any time if they feel uncomfortable. If participants initially agree and then decide to withdraw that agreement at the end of the study or after they have been debriefed, all data and information about them must be deleted from the research. Participates can withdraw themselves and their data.

Answer 70

Researchers have responsibility to protect the participants from harm. They must ensure that those taking part in research will not be caused distress. They must be protected from physical and mental harm. This means that you must not embarrass, upset, frighten, offend or harm participants. E.g. showing gruesome pictures could upset people; offend their culture; putting them in physical harm etc. … Psychologists need to consider whether the benefits of the research can be said to outweigh the possible costs to participants in that research. The risk of harm should be no greater that the risk from everyday life.

Answer 71

Psychologists should behave with honesty and fairness in all their interactions with people. Avoid deception, which means people should not be misled about the research. The problem is that deception is sometimes necessary to avoid demand characteristics affecting the results. Some research would be impossible if everything was revealed at the start. The guidelines say that participants must be deceived as little as possible, and that the deception must not cause distress. \* If you have gained participants’ informed consent by deception, then they will have agreed to take part without knowing what they were consenting to. The true nature of the research should be revealed at the earliest possible opportunity, or at least during debriefing. If serious deception has been used, the researcher must ensure that the participant understands why and feels comfortable about their part in it.

Answer 72

Before the study begins, the researcher must outline to the participant what the research entails, and then ask if they consent to take part. If the participants are children, someone who is responsible for them (parent/ guardian) must be told what is involved, and they must give consent. The researcher must still ask the child if they’re willing to take part, and the child must agree. However, it is not always possible to gain informed consent. This is acceptable as long as what happens to the participants is something that could just as easily have happened to them in everyday life. (E.g. observing people in a bus queue – they could be observed by anyone so informed consent is not necessary).

Answer 73

Participants must be thoroughly debriefed at the end of the study. They must be given a general idea of what the researcher was investigating and why, and their part in the research should be explained. They must be assured that their results are confidential. They must be told if they have been deceived and it must be justified to them. They must be asked if they have any questions and these should be answered honestly and as fully as possible. Participants may have experienced distress thorough their experience, perhaps why they hear they have been deceived or if the procedure caused them anxiety, embarrassment or loses of self-esteem. It is the researcher’s responsibility to check on the participants’ physical and psychological well-being as part of the debriefing process. If necessary they should be followed up to ensure there are no ill effects later on. Use our study and apply this information on ethical issues. EG look at the instructions we produced. Identity the ethical issues you need to consider and add these to your briefing, standardised instructions and debriefing.