Ethical issues
- Deception
- Consent
- Right to withdraw
- Protection from physical and psychological harm
- Dealing with ethical issues e.g. debrief, committees and guidelines
- Cost-benefit analysis
Research Methods
Experimental
|
Non experimental
|
Research Design and implementation
Data analysis
However, this is not a logical teaching order so although this booklet will cover all the stuff mentioned above it will follow a different sequence and hopefully one more similar to the route taken in class.
- Aims and Hypotheses
- Research design
- Independent, dependent and extraneous variables
- Sampling
- Pilot studies
- Reliability and validity
- Demand characteristics and investigator effects
Data analysis
- Analysis of quantitative data
- Measures of central tendency and of dispersion
- Correlation coefficients
- Presentation and interpretation of quantitative data
- Analysis and interpretation of qualitative data
- Presentation of qualitative data
However, this is not a logical teaching order so although this booklet will cover all the stuff mentioned above it will follow a different sequence and hopefully one more similar to the route taken in class.
Ethical issues in Psychological research
Ethics are the moral codes laid down by professional bodies to ensure that their members or representatives adhere to certain standards of behaviour. All scientific bodies have such codes but those in psychology are particularly important because of the subject matter of the topic.
- Psychology is unlike most other subject areas in that its subject matter is entirely human or animal. Because of this practically all research involves living things that can be caused physical or psychological harm.
- Psychological research also needs to consider the wider community. Milgram’s research taught us something unpleasant about the human race in general. Some research, for example studies on IQ, have been used to discriminate against different races or ethnic groups. It could be argued that Bowlby’s research was used to discriminate against women, making them feel guilty for not being at home caring for their children.
- The knowledge gained from psychological research can be exploited by people or groups to gain an advantage over others. Skinner’s work on behaviour shaping could be abused in this way.
Protecting the individual in psychological research
Many of the ideas mentioned in this section will be raised as we cover other topics later in the year and particularly in the last topic on social influence.
- Deception
- Consent (informed or not)
- Protection of participants from physical and psychological harm
- The right to withdraw
- The right to withdraw data
- Confidentiality and Privacy
Why Were Ethical Guidelines Introduced in Psychology? To protect the well being, health and dignity of participants in psychological research. Psychologists moved away from a view of people as “Subjects” in experiments – to a more humanistic approach to them as active “Participants” in research. Milgram’s procedure involved deception, lack of informed consent, physical and psychological harm, denied participants their confidentiality and right to withdraw (allegedly). However, a therapeutic debrief was provided and no ethical guidelines were broken since they didn’t exist at the time! Did what we learn justify these methods? |
Deception
Examples of studies involving deception: Asch, Milgram, Moscovici
Deception involves either concealing the real intention of a study from participants or taking steps to mislead them at the outset. All of the examples above used the second ploy, deliberately lying to participants about the genuine reason for a study. Two of them also used stooges or confederates (people pretending to be participants who are really part of the experimental set up). The use of stooges always means deception has been used.
However, is deception necessary? The researchers above would all argue that their experiments could not have taken place without it. Imagine if Milgram had said at the start, ‘Mr Wallace is really a stooge, who will scream a bit but will receive no shocks.’ The study would have told us nothing of interest and obedience would doubtless have been close to 100%.
To a lesser extent nearly all studies involve an element of deception in that it generally isn’t a good idea to tell your participants what you are looking for in advance. Menges (1973) estimated that as few as 3% of studies involve no deception at all. When using the BEM sex role inventory to test gender, telling male participants in advance that you are trying to find how masculine or feminine they are will almost certainly influence the way they respond to the questionnaire!
Baumrind on the other hand argues that deception is always wrong since it prevents informed consent (see below), researchers have an obligation to protect their participants (see below) and psychologists should be seen as professional and therefore trustworthy.
Debriefing
It is really a matter of common courtesy to debrief your participants at the end of any procedure and inform them of the point of the research. Debriefing is crucial if any form of deception has been employed.
A proper debrief should:
- Inform participants of the purpose of the research
- Ensure that there are no negative or unforeseen consequences of the procedure
- Ensure that the participant leaves in ‘a frame of mind that is at least as sound as when they entered.’ (Aronson 1988).
- Give the participant the right to withdraw their data and to see the finished write-up of the report if they so wish.
As well as having the best interests of the participant in mind, debriefs can also be a useful source of additional information in an experiment. Participants may tell you things that you would otherwise not be aware of.
Therapeutic debriefing
In extreme cases such as Zimbardo’s study, participants may receive questionnaires, be asked to complete diaries and have follow up meetings with the experimental team. In the case of Milgram some participants also received follow up psychiatric visits!
Consent and Informed consent
Consent
Simply refers to participants willingly and voluntarily taking part in your experiment. Milgram and Asch for example did obtain consent. In the case of Milgram he placed his infamous advert in the local paper and people turned up. During WWII the Nazis carried out many procedures on prisoners without their consent. Following the war it was decided that consent should be enshrined as a basic human right.
Informed consent
This refers to participants giving their consent in full knowledge of the aims of the study, the expectations of them and their right to withdraw and to confidentiality. This clearly was not the case with Asch or Milgram, but arguably was with the Zimbardo procedure. This raises the issue of whether fully informed consent is ever possible. If researchers know the likely outcomes of a study then what is the point in carrying it out in the first place?
Informed consent and deception are closely related in that there cannot be informed consent in any situation where deception is used.
Special cases
Children
Children under the age of 16 are deemed not to be old enough to give consent. In this case permission has to be sought from parents or guardians. Students It has been common practice by many universities to expect students to participate in experiments as a requirement of the course. In my fresher year I was expected to earn a certain number of points by being a participant in studies. Those involving pain (like the electric shocks I suffered in acquiring my aversion to the number 3) gained higher points. Here a certain degree of coercion is used and may not be entirely ethical. |
Detained
People in prisons or psychiatric hospitals need particular consideration. Prisoners may feel pressured into taking part as failing to do so may prejudice their situation. Similar concerns apply to patients. Additionally with psychiatric patients permission may need to be sought from either relatives or psychologists. Field experiment Piliavin conducted research on the NY underground in which stooges pretending to be blind or drunk (not both!), fell over. The research team observed the reactions of bystanders. In situations like this ‘participants’ are not aware that they are taking part in a study so cannot give consent. In addition it is usually impossible to carry out debriefs afterwards |
Protection from physical and psychological harm
Physical harm
The BPS guidelines suggest that participants should be exposed to no more risk than they would be in everyday life. For example people driving cars are exposed to a certain level of risk. If psychologists wish to study some aspect of driving related behaviour then the procedure they use should not put their participants at greater risk than this.
There are occasions when researchers have caused their participants physical harm although these tend to be rare. Milgram appears to have delighted in the response of some of his participants who would ‘bite their lips and dig their fingernails into their flesh. Full blown, uncontrollable seizures were experienced by three subjects.’ (Wrightman and Deux 1979).
Psychological harm
This is more difficult to gauge but may involve embarrassment, loss of self esteem, stress and anxiety.
Asch, Zimbardo and Milgram procedures would all have involved loss of self esteem, embarrassment and some stress, and in the case of Milgram and Zimbardo, extreme anxiety.
Confidentiality is one way of protecting participants from psychological harm. If you do something shameful or embarrassing then others not knowing will help reduce the impact.
Confidentiality
The data protection act requires that the identity of all participants remains confidential. As well as safeguarding privacy there is an obvious practical benefit from this approach. Participants are unlikely to volunteer for procedures if they believe that their identity and behaviour will be divulged.
There were clear breaches of confidentiality in the Milgram and Zimbardo studies as in both cases participants were secretly filmed.
Guidelines require that participants are not identified unless they give their permission and various methods may be used to disguise their identity. For example in case studies patients may be identified only by their initials such as KF or HM.
The right to withdraw and to withdraw data
This should be available and made clear to participants before the research starts. Both Milgram and Zimbardo claim that withdrawal was possible in their studies although when questioned afterwards it is clear that not all participants realised this.
Advance payment was an issue in the Milgram study. This may put additional pressure on participants who may feel obliged to earn the money that they have received. The debrief should make it clear that participants have the right to withdraw their data on being told the nature of the study. If serious deception has taken place then participants have the right to witness their data being destroyed!
Dealing with the ethical issues
This is a favourite question in which you are expected to describe and/or evaluate measures taken by psychologists to minimise the adverse effects of research. Obvious points to mention would be seeking consent, avoiding deception, providing the right to withdraw, debriefs and confidentiality.
For fuller marks some or all of the following could also be discussed:
Following the immoral experiments of the Nazis in WWII, each country set up its own set of guidelines for performing scientific research. In Britain the British Psychological Society (BPS) and in the USA the American Psychological Association (APA), produce codes of conduct for both experimentation and for clinical practice.
For human participants the codes cover topics already mentioned such as deception, consent, withdrawal of data, confidentiality etc.
Additionally all institutes that perform psychological research have ethical committees that consider whether or not particular pieces of research should be carried out. This body should have non psychologists that can express more objective views on research.
Obtaining consent and avoiding deception
Role playing
People are asked to act out the role of participants in problematical studies involving deception or psychological harm etc. Clearly these are less than satisfactory since people can only guess at how they would respond in such situations. When asked, fewer than 1% of people believe that they would obey in Milgram’s study!
Cost-benefit analysis and the Double Obligation Dilemma
Researchers are paid to carry out studies that uncover interesting and useful information about the human experience. As such they have an obligation to society and the tax payer that pays their wage. Psychologists also have an obligation to protect those who agree to take part in research from harm. Occasionally these obligations can come into conflict, as we saw with the Milgram and Zimbardo studies. In such cases a careful cost benefit analysis needs to be carried out prior to the commencement of research. Ethical committees need to carefully weigh up whether the costs incurred are worth the likely benefits gained. Trouble is… how can the benefits be assessed in advance? How can committees foresee the likely costs? Nobody believed Milgram’s participants would behave the way they did. Zimbardo expected his study to last two weeks. It had to be abandoned after six days.
Why Is It Important To Adhere To Ethical Guidelines?
To ensure that all participants are protected. This will give psychology a good name and should mean that participants will be willing to take part again in any future research. This is very important because participants are vital for psychological research. Any psychological harm suffered by participants would give psychology a bad name and would deter further participation in research.
Why Is It Important To Gain The Consent Of Participants In Psychological Research?
This helps psychologists because it protects them from participants who may wish to complain about their treatment in a study. If participants have given their consent, they do not have grounds to complain BUT it depends whether or not their consent was informed. Informed consent can be a bad thing because the participants may act differently if they know the purpose of the experiment therefore affecting the validity of the research.
Why Is It Helpful To Use Standardised Instructions In Psychological Research?
This ensures that all participants are treated ethically and in the same way and may therefore be viewed as a control which means that the study may easily be replicated. This in turn can boost the scientific status of psychology.
Why Is It Important For Participants To Trust Psychologists?
If participants trust psychologists they will view them as being professionals and this may help to boost the scientific status of psychology. Participants in research should have confidence in the investigators; this is achieved by treating them ethically according to the guidelines.
Why Is Debriefing Important?
Debriefing can be a good thing because it means that the participants understand the rationale for the research and it can provide them with closure but debriefing may also be a bad thing because lots of people who have participated in psychological research may share their experiences with others who also become aware of the details and therefore there may be a shortage of naïve participants in future replications of the study. Furthermore this could increase demand characteristics in future participants which could affect the validity of the study.
Physical harm
The BPS guidelines suggest that participants should be exposed to no more risk than they would be in everyday life. For example people driving cars are exposed to a certain level of risk. If psychologists wish to study some aspect of driving related behaviour then the procedure they use should not put their participants at greater risk than this.
There are occasions when researchers have caused their participants physical harm although these tend to be rare. Milgram appears to have delighted in the response of some of his participants who would ‘bite their lips and dig their fingernails into their flesh. Full blown, uncontrollable seizures were experienced by three subjects.’ (Wrightman and Deux 1979).
Psychological harm
This is more difficult to gauge but may involve embarrassment, loss of self esteem, stress and anxiety.
Asch, Zimbardo and Milgram procedures would all have involved loss of self esteem, embarrassment and some stress, and in the case of Milgram and Zimbardo, extreme anxiety.
Confidentiality is one way of protecting participants from psychological harm. If you do something shameful or embarrassing then others not knowing will help reduce the impact.
Confidentiality
The data protection act requires that the identity of all participants remains confidential. As well as safeguarding privacy there is an obvious practical benefit from this approach. Participants are unlikely to volunteer for procedures if they believe that their identity and behaviour will be divulged.
There were clear breaches of confidentiality in the Milgram and Zimbardo studies as in both cases participants were secretly filmed.
Guidelines require that participants are not identified unless they give their permission and various methods may be used to disguise their identity. For example in case studies patients may be identified only by their initials such as KF or HM.
The right to withdraw and to withdraw data
This should be available and made clear to participants before the research starts. Both Milgram and Zimbardo claim that withdrawal was possible in their studies although when questioned afterwards it is clear that not all participants realised this.
Advance payment was an issue in the Milgram study. This may put additional pressure on participants who may feel obliged to earn the money that they have received. The debrief should make it clear that participants have the right to withdraw their data on being told the nature of the study. If serious deception has taken place then participants have the right to witness their data being destroyed!
Dealing with the ethical issues
This is a favourite question in which you are expected to describe and/or evaluate measures taken by psychologists to minimise the adverse effects of research. Obvious points to mention would be seeking consent, avoiding deception, providing the right to withdraw, debriefs and confidentiality.
For fuller marks some or all of the following could also be discussed:
- Ethical guidelines and codes of conduct
- Cost-benefit analyses
- Ways of obtaining consent and avoiding deception
Following the immoral experiments of the Nazis in WWII, each country set up its own set of guidelines for performing scientific research. In Britain the British Psychological Society (BPS) and in the USA the American Psychological Association (APA), produce codes of conduct for both experimentation and for clinical practice.
For human participants the codes cover topics already mentioned such as deception, consent, withdrawal of data, confidentiality etc.
Additionally all institutes that perform psychological research have ethical committees that consider whether or not particular pieces of research should be carried out. This body should have non psychologists that can express more objective views on research.
Obtaining consent and avoiding deception
- Presumptive consent (of ‘reasonable people’)
- This asks people for their views on a particular procedure. If generally they find it acceptable then that procedure is used… but NOT on those asked.
- Prior general consent
- A pool of possible participants would be asked for their views on research. For example they may be asked about their views on the use of deception or embarrassment during research. Only those participants who consider these ploys acceptable would then be selected for later research involving fibs etc.
Role playing
People are asked to act out the role of participants in problematical studies involving deception or psychological harm etc. Clearly these are less than satisfactory since people can only guess at how they would respond in such situations. When asked, fewer than 1% of people believe that they would obey in Milgram’s study!
Cost-benefit analysis and the Double Obligation Dilemma
Researchers are paid to carry out studies that uncover interesting and useful information about the human experience. As such they have an obligation to society and the tax payer that pays their wage. Psychologists also have an obligation to protect those who agree to take part in research from harm. Occasionally these obligations can come into conflict, as we saw with the Milgram and Zimbardo studies. In such cases a careful cost benefit analysis needs to be carried out prior to the commencement of research. Ethical committees need to carefully weigh up whether the costs incurred are worth the likely benefits gained. Trouble is… how can the benefits be assessed in advance? How can committees foresee the likely costs? Nobody believed Milgram’s participants would behave the way they did. Zimbardo expected his study to last two weeks. It had to be abandoned after six days.
Why Is It Important To Adhere To Ethical Guidelines?
To ensure that all participants are protected. This will give psychology a good name and should mean that participants will be willing to take part again in any future research. This is very important because participants are vital for psychological research. Any psychological harm suffered by participants would give psychology a bad name and would deter further participation in research.
Why Is It Important To Gain The Consent Of Participants In Psychological Research?
This helps psychologists because it protects them from participants who may wish to complain about their treatment in a study. If participants have given their consent, they do not have grounds to complain BUT it depends whether or not their consent was informed. Informed consent can be a bad thing because the participants may act differently if they know the purpose of the experiment therefore affecting the validity of the research.
Why Is It Helpful To Use Standardised Instructions In Psychological Research?
This ensures that all participants are treated ethically and in the same way and may therefore be viewed as a control which means that the study may easily be replicated. This in turn can boost the scientific status of psychology.
Why Is It Important For Participants To Trust Psychologists?
If participants trust psychologists they will view them as being professionals and this may help to boost the scientific status of psychology. Participants in research should have confidence in the investigators; this is achieved by treating them ethically according to the guidelines.
Why Is Debriefing Important?
Debriefing can be a good thing because it means that the participants understand the rationale for the research and it can provide them with closure but debriefing may also be a bad thing because lots of people who have participated in psychological research may share their experiences with others who also become aware of the details and therefore there may be a shortage of naïve participants in future replications of the study. Furthermore this could increase demand characteristics in future participants which could affect the validity of the study.
Cost-benefit analyses
Committees may carry out cost-benefit analyses in which the likely benefits of a particular piece of research is weighed up against the costs to human or animal participants. Put simply does the knowledge we gain about human behaviour and the advantages this might have for the wider population warrant the suffering or embarrassment of a few individuals? Such analyses are notoriously difficult to carry out objectively, particularly in advance of a piece of research. Psychologists still argue about the costs and the benefits of the Milgram procedure, and that’s with the benefit of forty years of hindsight! Additionally the costs to the larger social group may also be considered, for example and an ethnic or racial group or women etc. |
How Might Ethical Guidelines Limit What Psychologists Can Study?
Most controversial and groundbreaking psychological research was carried out before the introduction of ethical guidelines and allowed the researchers to access completely raw human reaction to situations from completely naïve subjects. The introduction of the guidelines has placed restrictions upon such studies meaning that participants cannot be deceived and must be fully informed about the research. Consequently psychologists have investigated less intrusive methods of research, which do not rely on manipulation of behaviour, e.g. naturalistic observation, discourse analysis
Why Does The Guideline Of Deception Place Constraints On Research?
In some psychological research an element of deception is necessary in order to obtain ‘real’ results and prevent demand characteristics and also to help achieve validity in the results. The introduction of this guideline restricts psychologists’ ability to withhold information from participants.
Research Methods
The Experiment
In an experiment a variable is manipulated to see what effect it will have on another. For example if we wanted to know whether caffeine affected reaction times:
We could take two groups, give one group coffee (experimental group) and compare them to another group without coffee (control group). We would then set them a task designed to measure their reaction times.
In experiments there are 2 variables:
- Independent variable (the one we alter or manipulate) in this case whether or not the person has had coffee.
- Dependent variable (the one that alters as a result of what we do), in this case reaction time. The dependent variable is usually the one we measure or record.Not rocket science, BUT the problem is always remembering which is which. The way I do it is simply to think of the dependent variable as the one that depends on what we do!
In this case reaction time (dependent variable) depends on whether or not the participant has had a cup of coffee.
Crucially an experiment allows us to establish a causal link between the IV and the DV. Following an experimental procedure we should be certain that the alteration we have made in the IV has caused the change in the DV.
Other variables
The I.V. and D.V., as we psychologists refer to them, are not the only variables to worry about.
In the coffee experiment suppose we find that the coffee group have faster reaction times, can we be certain the coffee has caused this. Other possible reasons:
- The experimental group (on coffee) might just by chance contain people with faster reactions
- Perhaps we measured one group in the morning, the other in the afternoon.
- Perhaps those in the control group had a hangover etc.Confounding variables
These are variables that get in the way of our results or make our results difficult to interpret.
Common confounding variables include:
Intelligence of participants, Personality of participants, Gender of participants, Time of day, Weather, Noise levels, Temperature…
Obviously in an experiment we take steps to minimise these, for example we could ensure that the procedure is carried out at the same time of day, in the same room, with similar temperature settings etc.
Laboratory experiments
Lab experiments don’t have to be carried out in a laboratory. However, any experiment that is carried out in a special, tightly controlled environment is classed as laboratory. Importantly it is obvious to those taking part that that they are in an experimental procedure. Laboratory experiments are therefore artificial and tightly controlled, leading to the following advantages and disadvantages:
Advantages of lab experiments
Cause and effect: We can usually see that the IV has caused the alteration in the DV. If we have controlled our experiment we should be able to show that it was the coffee that was responsible for the faster reaction times. Replication: Provided care has been taken in conducting and reporting the procedure another person should be able to repeat your procedure to see if they get the same results. |
Disadvantages of lab experiments
Lacks ecological validity: As we’ve seen so many times (e.g. in memory and in Milgram), experiments, especially those in laboratories are very artificial. Can they really tell us how people will behave in real life situations? Demand characteristics: New one for you; this refers to participants behaving differently because they know they’re being watched. We saw this in Milgram. It could be that they guess what you want and try to please the experimenter, e.g. by obeying! |
Not all experiments however are carried out in artificial settings and not all allow full control of the IV. Other types of experiment are covered next:
Field experiments
Not, as the name implies, experiments conducted in fields, although they could be! More likely settings would include the work place, school, the street etc. Basically the same rules apply: an independent variable is manipulated to see how it affects a dependent variable. Confounding variables can still get in the way, and cause and effect can still be determined. However, the setting is more natural.
Field experiments
Not, as the name implies, experiments conducted in fields, although they could be! More likely settings would include the work place, school, the street etc. Basically the same rules apply: an independent variable is manipulated to see how it affects a dependent variable. Confounding variables can still get in the way, and cause and effect can still be determined. However, the setting is more natural.
Advantages of field experiments
Ecological validity: because the settings are more natural it is assumed that people will behave more naturally, so field experiments should have greater ecological validity. Demand characteristics: these can be less since participants may not be aware that they are in an experiment, as was the case with Hofling! |
Disadvantages of field experiments
Less control of variables: the experimenter has less control over the environment so more variables may affect the outcome. As a result we cannot be certain that the IV has caused the change in DV. Ethics: If patients are unaware of the study how can the consent to take part or withdraw from the experiment? Replication: It is difficult to repeat the procedure exactly as it was the first time. |
In practice it can be difficult to distinguish a laboratory experiment from a field experiment. Consider the Brewer and Treyen’s office schema study. This is clearly an experiment and in an artificial setting, but the participants were not aware at the time of the procedure, that they were taking part in a study so there behaviour was quite natural. Similarly with Loftus’ study on weapons focus. Participants listened to an argument whilst waiting to start an experiment. Again the setting was artificial and there was full control over the IV (blood soaked knife or pen). However, again the participants were unaware of the procedure so again their response would have been natural. Are these laboratory experiments or field experiments? No right answers really. However, the crucial thing is that you can justify your answer and explain the positive and negative points.
Natural experiments
Not, as the name implies, experiments carried out in the buff, although they could be if you were comparing the memory of those who naturally prefer to go au naturelle with those who prefer to wear clothes. These are similar to and often confused with quasi-experiments, but there is one crucial difference. Natural experiments take advantage of a naturally occurring event. The effect of the eruption of Mount St Helens on stress related illnesses is the one all the texts prefer to mention. In this case the IV was the eruption, a naturally occurring event.
A better example and one that we’ve studied is Hodges & Tizard’s study of institutional care which examined the effect of different types and duration of care on the children’s subsequent behaviour and development.
IV is the type and duration of care (in this case not controlled by the researchers, it happened anyway). DV is the effect this has on subsequent development (and which can be measured using various tests).
Natural experiments
Not, as the name implies, experiments carried out in the buff, although they could be if you were comparing the memory of those who naturally prefer to go au naturelle with those who prefer to wear clothes. These are similar to and often confused with quasi-experiments, but there is one crucial difference. Natural experiments take advantage of a naturally occurring event. The effect of the eruption of Mount St Helens on stress related illnesses is the one all the texts prefer to mention. In this case the IV was the eruption, a naturally occurring event.
A better example and one that we’ve studied is Hodges & Tizard’s study of institutional care which examined the effect of different types and duration of care on the children’s subsequent behaviour and development.
IV is the type and duration of care (in this case not controlled by the researchers, it happened anyway). DV is the effect this has on subsequent development (and which can be measured using various tests).
Advantages of natural experiments
Demand characteristics: it is often the case that the experimenter isn’t even present when the event occurs, thankfully in the case of Mt Saint Helens! As a result participants are not trying to please the researchers. Research opportunities: it is possible to research events that it would be unethical to study any other way or that may be impossible to set up. |
Disadvantages of natural experiments
Lack of control: the researchers have no control at all over the variables and there may be lots of confounding variables. In the Mt St Helens case ill health caused by smoke, or stress due to loss of house etc. Replication: in some cases clearly impossible, in others very difficult. As a result it may be impossible to check the validity of research. Cause and effect: following on from lack of control, it may be impossible to decide if the IV is causing the change in the DV. |
Quasi-experiments (or tests of difference)
Not, as the name implies, experiments were you run around in the dark pretending to shoot each other with lasers! But they could be if you were looking for age or sex differences.
In a real experiment you can manipulate the IV and you can decide who goes in which group. In your study on coffee you decide which participants go in which group. However supposing you wanted to see if 40-somethings had faster reactions than teenagers: you never know it could happen!
In one group your participants will have to be teenagers and the other group will have to comprise 40-somethings. You are unable to randomly allocate your participants to the different groups. Similarly with sex differences; by definition the boys are going to be in one group and the girls in the other!
Experimental research design
Here we decide how we are going to sort or group our participants. Do we use the same people in all conditions or groups, or do we choose different people for different conditions or groups? In some cases, as we’ll see the decision is made for us. In others the solution isn’t so obvious and there may be pros and cons for each.
Repeated Measures Design
Here we use the same participants in each group or condition.
For example, returning to the earlier experiment on coffee and reaction times.
In a repeated measures design we could give our group of participants the test on day one with no coffee and record their reaction times.
The next day we could repeat the procedure, with the same group of people, but this time give them coffee before the experiment began.
Advantages
The two groups have the same age, sex, personality, ideas, past experiences, IQ, reaction times (crucially for this one) etc. They are perfectly matched. They are the same people!
Disadvantage
Order effects: Assuming, as we expect the group do better on the second day, can we be sure that this increase in performance is due to the coffee? It could be that they’ve had the chance to practice the task the day before! It’s not surprising they’re better the second time around. This is called order or practice effect.
Boredom: Of course, on some tasks it could work the other way, and a task done the second time shows a deterioration because they’re fed up with doing it.
Extra materials: For example if you use the same participants for two memory experiments you will need two lists of words etc. for them to recall. This introduces other variables. Perhaps the second list is easier than the first.
Guessing the point: Carrying out a procedure twice makes it more likely that participants will guess the aim of the research.
Counterbalancing
To overcome order or boredom effects we could use ABBA. One half of the participants could do no coffee followed by coffee the next day (condition A followed by condition B). The other half could do coffee on the first day and no coffee on the next day (condition B followed by condition A). Hence ABBA (nothing to do with thanking anyone for the music!). In some cases repeated measures has to be used: If you’re comparing gender with subject choice at AS you use the same people in each condition and compare a persons gender score with their AS choices. Examples we’ve seen this year The love quiz: the same people take each questionnaire The 44 thieves: later behaviour is compared to early attachments in same people |
Independent Measures Design
You guessed it. If we used the same people in each group last time, this time we use different people in each group. Clearly this overcomes practice and boredom effects ‘cos they only do it the once! Each participant is randomly allocated to one group or the other, so in our coffee experiment:
One group, comprising one set of participants do the test with coffee
The other group, comprising a different set of participants do the test without coffee.
Sorted, no problems with practice or repeat effects or with boredom or tiredness effects.
However
Can we be certain that the likely faster reactions of the first group are down to the coffee?
It could be that the participants that we’ve randomly assigned to that condition have naturally faster reactions. They may be younger, or some of them may engage in activities that require fast reactions.
In other experiments, sex, personality, age, IQ etc. could all be an issue because the participants are going to differ on all of these.
There are some occasions when independent measures design has to be used:
Sex differences
Age differences
By definition the two conditions are different. You couldn’t have someone in the male condition and the female condition, or in the under 30 condition and the over 30 condition!
Advantages
- No order or practice effects
- Can use the same stimulus material (such as word lists in memory) for each group
- Participants are not matched in terms of IQ, personality, age etc.
- You will need twice as many participants.
Matched Pairs Design
This is the ideal compromise. In the reactions experiment you would have different people in each condition, i.e. some would have the coffee and others not. However, the two sets would be matched in terms of IQ or whatever characteristics are relevant, in this case reaction times, age etc.
Advantages
- No order effects since each participant only does the task once
- You can use the same material twice
- Groups are similar in terms of individual characteristics
Disadvantages
- Very time consuming and difficult to match all of your participants in this way
- It is impossible to match people for all characteristics even if you were to use MZ twins between the two groups!
Observations
These are a vital tool in the psychologists armoury and if done properly can provide oodles of ecologically valid and detailed information about all manner of behaviours. However, they also have many pitfalls and raise a whole host of methodological and ethical issues. Observations can be subdivided in many ways each with their own distinct advantages and disadvantages. In unit one we have seen a few worthy examples of observations including the strange situation, Lorenz’ work on geese and in addition to these you should also be familiar with Bandura’s bobo doll research.
The following divisions of observations will be considered:
- Naturalistic v controlled
- Structured v unstructured
- Participant v non-participant
- Disclosed v undisclosed
Naturalistic Observation
This is an easy one to explain. People or animals are observed in their natural environment, without any sort of intervention or manipulation of variables and without their knowledge.
Examples include:
- Seyfarth & Cheney’s research on the warning calls of the vervet monkey
- Sylva’s study of play in young children.
- Much of the work carried out by Konrad Lorenz
Perhaps it was the prospect of
nesting in that fulsome beard or maybe the aromatic allure of his rough
shag. Either way, Lorenz was still
attracting the birds well into his seventies.
Humans can also be observed in natural situations, although this can have ethical implications. For example Laud Humphrey’s outrageous observation of gay men in public toilets! Posing as lookout or ‘watch queen’ he observed the practice of ‘tea rooming’ and gained addresses from a police insider after acquiring the number plates of those involved. He then gained confidential information about them by interviewing them under false pretences. |
The researcher observes behaviour in its natural environment as many of the ethologists studying animal behaviour record their information. Ainsworth’s study of attachments in Ugandan women would be a human example of naturalistic observation
Advantages of naturalistic observation Ecological validity: Clearly this provides data that is very high in ecological validity since it has not been tainted by observer intervention with the observed not usually knowing that their behaviour is being watched. Demand characteristics: For the same reason there should be no demand characteristics. If you’re not aware that you’re being observed then you won’t be trying to please the researcher. Detailed: Information collected tends to be more detailed and provides a fuller idea of behaviour than the sort of information that can be collected in a laboratory. Think of the criticisms of behaviour in the strange situation Sometimes this is the only possible way of doing research, especially if people are unwilling or unable to complete questionnaires or interviews. |
Disadvantages of naturalistic observation Reliability: there is the issue of bias. For example if a researcher is looking at aggressive acts in a football game and assumes that boys are going to be more aggressive, the results may inadvertently be interpreted in this way. Ethics: are a major problem with many observational studies and especially with naturalistic. Not knowing you’re being watched creates issues with privacy and participants not consenting to take part Cause and effect: However control of the environment is not possible and confounding variables make it impossible to determine cause and effect relationships. You cannot be certain what factors are creating the behaviour being observed Replication: in many cases it would be impossible to recreate exactly the same situation so that someone else could verify your findings. |
Controlled
As the name suggests the researcher in some way manipulates the behaviour of the observers or the observed. Ainsworth’s strange situation is the best example seen to date with researchers organising the behaviour of the mother and stranger to see how the child reacts. Other examples include the Bobo dolls and Piliavin’s work on bystander apathy on the New York subway.
These allow for greater control of confounding variables meaning it is easier to establish cause and effect relationships.
However, they are lower in ecological validity since the trigger for the behaviour is usually not a natural event. Often, but not always, participants may also know they are being observed creating demand characteristics.
Disclosed
Participants know they are being observed. This reduces ethical issues of consent and privacy but reduces validity due to increased demand characteristics.
Undisclosed
Participants are unaware of the observation. This raises ethical issues (privacy and consent) but increases validity by reducing demand characteristics. Sometimes one way mirrors might be used to discretely observe people, for example shopping behaviour in a supermarket.
Participant
Here the researchers get involved with the group of participants they are observing. Festinger (1956) joined a cult to observe how they would react when their predicted end of the World deadline came and went. The cult leader was able to reassure his flock that their prayers had saved the planet! This is an example of undisclosed participant observation. On occasions researchers may join in but make others aware of their role as psychologists.
On occasions researchers have been able to infiltrate groups and remain members for a period of time allowing for detailed, longitudinal information to be gathered, for example about the behaviour and motivations of street gangs and religious cults. It is difficult to see how such groups could be studied in any other way.
Clearly there are ethical issues with this type of deceitful observation and the researcher themselves may unwittingly interfere with the group dynamics and the behaviour of the group.
Non-participant
The more likely scenario in which participants are observed from a distance rather than the researchers infiltrating the group.
Ethics of observations
Observations raise a number of unique ethical issues. These vary depending on the nature of the observation taking place but here are a few:
Consent: participants are often unaware of being observed so have no opportunity to consent to taking part in your research.
Debrief: often there is no opportunity for a debrief. For example in Piliavin’s observation of bystander apathy on the New York subway, participants were observed without their knowledge and would have left the train before researchers had chance to debrief.
Deception: participants being unaware of observation is deception in itself. Additionally, researchers may cause additional deception by using stooges. Again Piliavin used members of the research team to pretend to be blind or drunk.
As the name suggests the researcher in some way manipulates the behaviour of the observers or the observed. Ainsworth’s strange situation is the best example seen to date with researchers organising the behaviour of the mother and stranger to see how the child reacts. Other examples include the Bobo dolls and Piliavin’s work on bystander apathy on the New York subway.
These allow for greater control of confounding variables meaning it is easier to establish cause and effect relationships.
However, they are lower in ecological validity since the trigger for the behaviour is usually not a natural event. Often, but not always, participants may also know they are being observed creating demand characteristics.
Disclosed
Participants know they are being observed. This reduces ethical issues of consent and privacy but reduces validity due to increased demand characteristics.
Undisclosed
Participants are unaware of the observation. This raises ethical issues (privacy and consent) but increases validity by reducing demand characteristics. Sometimes one way mirrors might be used to discretely observe people, for example shopping behaviour in a supermarket.
Participant
Here the researchers get involved with the group of participants they are observing. Festinger (1956) joined a cult to observe how they would react when their predicted end of the World deadline came and went. The cult leader was able to reassure his flock that their prayers had saved the planet! This is an example of undisclosed participant observation. On occasions researchers may join in but make others aware of their role as psychologists.
On occasions researchers have been able to infiltrate groups and remain members for a period of time allowing for detailed, longitudinal information to be gathered, for example about the behaviour and motivations of street gangs and religious cults. It is difficult to see how such groups could be studied in any other way.
Clearly there are ethical issues with this type of deceitful observation and the researcher themselves may unwittingly interfere with the group dynamics and the behaviour of the group.
Non-participant
The more likely scenario in which participants are observed from a distance rather than the researchers infiltrating the group.
Ethics of observations
Observations raise a number of unique ethical issues. These vary depending on the nature of the observation taking place but here are a few:
Consent: participants are often unaware of being observed so have no opportunity to consent to taking part in your research.
Debrief: often there is no opportunity for a debrief. For example in Piliavin’s observation of bystander apathy on the New York subway, participants were observed without their knowledge and would have left the train before researchers had chance to debrief.
Deception: participants being unaware of observation is deception in itself. Additionally, researchers may cause additional deception by using stooges. Again Piliavin used members of the research team to pretend to be blind or drunk.
Correlational analysis
In the past year we’ve seen lots of examples of this. For example whenever I’ve criticised a study because it doesn’t show cause and effect it’s probably been a correlational study.
An example:
For example we could look for a correlation between IQ and performance at GCSE or A-level. Common sense would perhaps tell us that students that have higher IQs are more likely to perform well at GCSE.
In year 13 we look at the controversial area of IQ and find that there is a high correlation between the IQs of MZ twins. If one twin has a high IQ it is likely the other does too. This is taken as evidence for the nature or genetic determination of IQ. However, as we will see there are a host of other reasons why this might be the case.
Types of correlation
Positive: the most common; as one variable increases so does the other, e.g. IQ and GCSE score in the example above.
Negative: as one variable increases the other decreases, e.g. it might be fair to assume that the higher your stress levels the lower your life expectancy. Again we are unable to show cause and effect. As mentioned frequently in ‘Stress,’ illnesses could be due to secondary habits such as smoking, poor diet etc.
Advantages of correlations
Correlations allow us to study links between variables that could not be studied in any other way. We could not inflict so much stress on a person that we endanger their life. However, we can use a correlational analysis to show a possible link between the two occurring naturally. Economical and fast: large amounts of data can be compared quickly and cheaply, e.g. by using a questionnaire to collect data. |
Disadvantages of correlations
Cause and effect: do I really need to explain this one? A correlation shows a possible link between 2 variables it does not prove that one causes the other, e.g. smoking and heart disease. Correlations can disadvantage certain people in society if misused. For example it was established long ago that blacks under perform on IQ tests compared to whites. This knowledge was misinterpreted as evidence of white superiority. |
Case studies
These involve study of an individual, small group, institution or an event. A case study can involve a whole host of techniques including observations, questionnaires, surveys, interviews, testing and even on occasion experiments. They are frequently longitudinal in nature and may also involve asking others, such as friends and associates.
Examples:
Clive Wearing, HM, KF, S, Genie, Czech twins, Anna O, Little Albert, Little Hans, Phineas Gage
Good points
They provide a wide variety of in-depth and detailed information that would be impossible to acquire using heavily controlled situations such as experiments. They can offer provide a real feel for what it is like to be suffering from a particular disorder or be involved in a certain situation.
They often provide the only method possible of studying a certain condition or event. It would not be possible to artificially re-create situations such as Genie or HM experimentally, so our only access to information about privation or severe amnesia is through case studies.
However
It is notoriously difficult to generalise from a case study and create a general theory. Case studies by their very nature are one-offs or unusual and often involve people who are not themselves representative of the general population. The case of Genie and the Czech twins shows this nicely. Both suffered severe deprivation over a prolonged period but their outcomes are very different; the Czech twins seeming to make a full recovery, whereas, as far as we know, Genie never recovered from her early problems.
Often case studies require an element of retrospective data collection, with parents, friends etc being asked to think back to the participants earlier years. Retrospective data collection is not reliable.
Objectivity by the researchers can be difficult, with psychologists getting too close to patients as with the case of David Rigler and Jean Butler and their research/fostering of Genie.
Confidentiality can be an issue though some of this can be overcome by the use of pseudonyms or initials.
Self Report Techniques
Interviews
There are a number of species of interview each with their own advantages and disadvantages. I’ll consider the main ones only:
Informal interviews
The interviewer has an aim in mind at the outset but is willing to be flexible about getting answers. The interviewer tries not to direct the interviewee but instead listens and lets the interview take its natural course.
Advantage
Disadvantages
Clinical interview
These were made popular by Freud and in particular Piaget and are a type of informal interview. Piaget for example would read ‘moral stories’ to a child and start off by asking the same questions to all the children, for example ‘who is the naughtier boy in the stories.’ However, follow up questions would be informal and vary from child to child.
Structured or formal interviews
These follow a set pattern with the interviewer having prepared a set of questions in advance that are asked in a particular order.
Note: sometimes the questions may be open and allow the interviewee to respond how they like, for example ‘how did you feel when Freddie ate your pet hamster?’ Or they can be closed and allow only a ‘yes’ or ‘no’ response. For example ‘were you upset when Freddie ate your pet hamster?’
Advantages
Limitation of interviews in general:
Social desirability bias
We all like to create a favourable impression. When faced with an interviewer we are less likely to be honest than when filling out an anonymous questionnaire. For example people being questioned about their love life are likely to exaggerate in face to face interviews.
Lie scales can be introduced to assess how honest answers may be. For example if people were being questioned about their childhood a ‘lie question’ might be; ‘As a child did you always do as you were told first time and without moaning?’ A response of ‘yes’ would be assumed to be a fib and indicate that perhaps the interviewees answers may not be reliable.
Questionnaires
We all know what they are and have all filled lots of them in. Basically a questionnaire is a list of written questions that is able to gather lots of relevant information relatively quickly and cheaply.
The biggest problem is wording of the questions. Again there is the issue of ‘open’ or ‘closed’, but more importantly, as we saw in EWT, the issue of leading questions. These are a favourite of politicians or of newspapers that want to find support or criticism of a particular issue. For example imagine you wanted to find out if people wanted more money spent on the NHS, a relatively neutral question might be:
These involve study of an individual, small group, institution or an event. A case study can involve a whole host of techniques including observations, questionnaires, surveys, interviews, testing and even on occasion experiments. They are frequently longitudinal in nature and may also involve asking others, such as friends and associates.
Examples:
Clive Wearing, HM, KF, S, Genie, Czech twins, Anna O, Little Albert, Little Hans, Phineas Gage
Good points
They provide a wide variety of in-depth and detailed information that would be impossible to acquire using heavily controlled situations such as experiments. They can offer provide a real feel for what it is like to be suffering from a particular disorder or be involved in a certain situation.
They often provide the only method possible of studying a certain condition or event. It would not be possible to artificially re-create situations such as Genie or HM experimentally, so our only access to information about privation or severe amnesia is through case studies.
However
It is notoriously difficult to generalise from a case study and create a general theory. Case studies by their very nature are one-offs or unusual and often involve people who are not themselves representative of the general population. The case of Genie and the Czech twins shows this nicely. Both suffered severe deprivation over a prolonged period but their outcomes are very different; the Czech twins seeming to make a full recovery, whereas, as far as we know, Genie never recovered from her early problems.
Often case studies require an element of retrospective data collection, with parents, friends etc being asked to think back to the participants earlier years. Retrospective data collection is not reliable.
Objectivity by the researchers can be difficult, with psychologists getting too close to patients as with the case of David Rigler and Jean Butler and their research/fostering of Genie.
Confidentiality can be an issue though some of this can be overcome by the use of pseudonyms or initials.
Self Report Techniques
Interviews
There are a number of species of interview each with their own advantages and disadvantages. I’ll consider the main ones only:
Informal interviews
The interviewer has an aim in mind at the outset but is willing to be flexible about getting answers. The interviewer tries not to direct the interviewee but instead listens and lets the interview take its natural course.
Advantage
- Lots of information can be gathered
- Interviewee made to feel relaxed
Disadvantages
- Difficult to analyse, especially if different participants discuss different issues
- Low reliability
Clinical interview
These were made popular by Freud and in particular Piaget and are a type of informal interview. Piaget for example would read ‘moral stories’ to a child and start off by asking the same questions to all the children, for example ‘who is the naughtier boy in the stories.’ However, follow up questions would be informal and vary from child to child.
Structured or formal interviews
These follow a set pattern with the interviewer having prepared a set of questions in advance that are asked in a particular order.
Note: sometimes the questions may be open and allow the interviewee to respond how they like, for example ‘how did you feel when Freddie ate your pet hamster?’ Or they can be closed and allow only a ‘yes’ or ‘no’ response. For example ‘were you upset when Freddie ate your pet hamster?’
Advantages
- Easily replicated
- Data is easier to analyse
- Data is less likely to be influenced by the interviewer
- Little flexibility so important points may be missed
- Questions may be ambiguous (think of the SRRS for determining stress levels).
- This format may encourage brief answers
Limitation of interviews in general:
Social desirability bias
We all like to create a favourable impression. When faced with an interviewer we are less likely to be honest than when filling out an anonymous questionnaire. For example people being questioned about their love life are likely to exaggerate in face to face interviews.
Lie scales can be introduced to assess how honest answers may be. For example if people were being questioned about their childhood a ‘lie question’ might be; ‘As a child did you always do as you were told first time and without moaning?’ A response of ‘yes’ would be assumed to be a fib and indicate that perhaps the interviewees answers may not be reliable.
Questionnaires
We all know what they are and have all filled lots of them in. Basically a questionnaire is a list of written questions that is able to gather lots of relevant information relatively quickly and cheaply.
The biggest problem is wording of the questions. Again there is the issue of ‘open’ or ‘closed’, but more importantly, as we saw in EWT, the issue of leading questions. These are a favourite of politicians or of newspapers that want to find support or criticism of a particular issue. For example imagine you wanted to find out if people wanted more money spent on the NHS, a relatively neutral question might be:
‘Should more money be spent on the NHS?’ The Mirror (presumably wanting a ‘yes’ response might get their pollsters to ask: ‘Should extra money be provided to the NHS to take care of Britain’s sick and elderly?’ Whereas the Telegraph (being very stereotypical here) may get their pollsters to ask: ‘Would you be happy to pay more taxes to fund bureaucracy in the NHS?’ Sir Humphrey explains how to get the perfect balanced sample :) |
|
Rather extreme examples admittedly, real surveys carried out by experienced pollsters would be far more subtle, but you get the idea!
It is always a good idea to test your questionnaire in a pilot study first to make sure it doesn’t take hours to complete and that participants understand the questions. Feedback like this may provide ideas for follow up questions to be asked in the real study.
Advantages
Disadvantages
Content analysis
Content analysis is basically an observation. However, it studies human behaviour indirectly usually by observing the things we produce, e.g. television programmes, magazines, web sites, TV advertisements etc.
An analysis of what we produce should be able to tell us a lot about the way we structure our society and about our values, prejudices and so on. For example a content analysis of television advertisements of the 1970s would probably paint a far more sexist view of the World than that present today, certainly in the UK at least.
Content analysis, though it often analyses written words, is a quantitative method, it produces numbers and percentages. After doing a content analysis, you might make a statement such as "27% of programs on HFM in May 2011 mentioned at Lady GaGa compared to only 8% in 2009.”
Though it may seem crude and simplistic to make such statements, the counting serves two purposes:
· to make analysis more objective
· to simplify the detection of trends.
What constitutes content?
All content is something that people have created. We call these artefacts. You can’t do content analysis of the weather - but if somebody writes a report predicting the weather, you can do a content analysis of that. Examples could include:
Newspaper items, magazine articles, books, catalogues, web pages, advertisements, graffiti, radio programs, news items, TV programs, photos, videos, films, music, speeches, interviews, plays, gestures, products in shops and as we saw in one question, drawings
Categories and Themes
Crucial to the whole process is categorising the data. For example if you were looking at paintings produced by patients with schizophrenia you may want to categorise in terms of colour, vividness, intensity, content etc. It is possible that you will have prior expectations and might opt for a top-down approach, having decided in advance what categories to look for. More likely you would use a bottom-up method in which your categories emerge from the material produced
The Process
The sample will be the artefacts that are to be analysed. This needs to be representative. For example, if looking at gender stereotypes in car adverts that appear in magazines, you would need a wide range of different magazines to get a representative sample. If you only used men’s magazines, your sample would be biased, and you may not be able to generalise your results.
Coding System: Similarly to an observation, the researcher has to create a coding system, which breaks down the information into categories. So for the example above, for each advert, you may first identify the gender of person in the car advert, and then tally what they are depicted doing. These behavioural categories might be:
Driving the car ·
The researcher would then tally each time either a man or woman was seen doing a particular behaviour in the advert. This is called a quantitative analysis.
An alternative to having a coding system like above is to do a qualitative analysis. This is where the researcher has categories and chooses a particular example to illustrate this category. So for the category “Driving the car” he might choose the picture above left as a demonstration. Instead of counting the data, it is described (hence qualitative rather than quantitative).
Results: the researcher then looks at the data he has collected, and draws conclusions. For example, he may find that many more men are depicted driving the cars than women, and women are more likely to be seen as passengers. He might conclude from this that there is a gender bias in the way cars are advertised.
Strengths and weaknesses of content analysis
An example
Manstead and McCulloch (1981) watched 170 television advertisements in a week and scored them on a whole range of factors such as gender of product user, gender of person in authority, gender of person providing the technical information about the product and so on.
Dealing with validity in content analyses
As in other research methods, validity can be affected by having an unrepresentative sample. Remember, in a content analysis, the sample is not a group of people, but the artefacts that you decide to analyse.
To deal with this, the researcher needs to ensure that the sample is representative. For example, if looking for racial stereotypes in TV adverts, you would need a range of adverts, for a range of different products, which are shown at different times of the day and on different channels.
The coding system could also be an issue with regards to validity. The coding system may not actually be measuring what you intend to measure. We can assess how valid our measuring system by:
Content validity: are we measuring what we intent to measure? We could ask a panel of experts of assess our coding system.
In a content analysis, there may be an observer bias. The observer doing the analysis may have an idea of what they hope to find, and so have a bias to only record those bits of information that fit their theory.
One way to overcome this is to use a double blind technique, where the person doing the observing does not know that aim of the study, and so will not have a bias.
Dealing with reliability in content analyses
The reliability of a content analysis refers to how consistent we would expect the results to be. In other words, the same results should be gained if repeated. The coding system should therefore be clear and easy to use. It should be objective so that a particular behaviour will only be recorded in one category. There are a number of ways to assess reliability in a content analysis:
Test-retest: another researcher (or even the same researcher) can repeat the analysis using the same coding system and the same artefacts. If the analysis is reliable, the same results should be gained.
Inter-observer reliability: Two or more observers can analyse the same artefacts. Their results are correlated. If the coding system is reliable, there should be more than 80% agreement.
Reliability can also be increased by training the observers in the use of the coding system through practice.
Research Design and Implementation
Aims and Hypotheses
Aims
When carrying out a piece of research it is essential that you have an aim in mind. This needs to be reasonably precise, for example ‘I’m gonna study memory’ would not be sufficiently precise. However the aims are broader, or less precise than the hypotheses. A suitable aim for memory might be ‘to see if age affects the duration of STM.’
Miller’s aim was to discover the capacity of STM
Hypotheses
Psychologists start with a theory which is a general idea about a behaviour and then develop a hypothesis which makes the theory testable. Hypotheses are more precise and should be operationalised, i.e. give some clue as to how the research will be carried out.
Experimental or alternative hypothesis
This makes your prediction, for example:
‘As age increases the duration of STM decreases.’
Directional or non-directional?
Having decided on your hypothesis and aims you need to decide on the direction. In the examples above I’ve already done this. When we say that we expect ‘duration of STM to decrease as age increases’ we are making a definite prediction. That prediction has direction. Compare this to the statement that ‘duration of STM will be affected by an increase in age.’ Will duration increase or decrease? The hypothesis doesn’t say. It could go either way.
Directional
If the hypothesis has a direction we say it is ‘directional’ or one-tailed. In the first example we are saying that duration of STM will decrease.
Non-directional
If we are not prepared to commit ourselves and simply say there will be an affect then this is non directional or two tailed.
Exam tips
If you’re asked to construct a hypothesis try to be specific. Consider the IV and DV or the covariables (in the case of a correlational analysis). If the DV is memory indicate how ‘memory’ is being assessed. For example ‘alcohol will decrease the number of words correctly recalled.’ This is essential if the question asks you to operationalise your hypothesis.
If you’re asked to justify why a directional or non-directional hypothesis has been chosen, the answer should be simple. Look at the stimulus material. If no prior research is mentioned then opt for two-tailed. The researchers are unlikely to be sure about what will happen so they’re going to edge their bets. Previous research that conclusively points to a certain expectation and you’ll opt for direction (in the direction of those previous findings). Justify your choice in terms of previous research (or lack of).
For example:
Psychological research has indicated that there is a gender difference in the way individuals behave at controlled pedestrian crossings (where the crossing is controlled by a red and green light). It has been found that females are more likely than males to comply with the lights.
If asked to produce a hypothesis and justify your choice of directional or non-directional;
Males are more likely to display independent behaviour at a controlled pedestrian crossing than females.
Directional has been chosen since previous research suggests women are more likely to comply at pedestrian crossings.
If the study is correlational then use the word correlation in your hypothesis.
If you are opting to go non-directional then ‘correlation’ is sufficient. If you’re opting for directional then you will need to precede with ‘positive’ or ‘negative.’
There will be a negative correlation between age and duration of STM as determined by the B-P technique.
It is always a good idea to test your questionnaire in a pilot study first to make sure it doesn’t take hours to complete and that participants understand the questions. Feedback like this may provide ideas for follow up questions to be asked in the real study.
Advantages
- Lots of people can be tested quickly
- This allows more reliable generalisation to the overall population
- Data can often be analysed easily
Disadvantages
- Lots of questionnaires will not be returned!
- People may tell fibs. Even in anonymous questionnaires this may be an issue. Again lie questions may be included, e.g. in Eysenck’s Personality Questionnaire (EPQ).
Content analysis
Content analysis is basically an observation. However, it studies human behaviour indirectly usually by observing the things we produce, e.g. television programmes, magazines, web sites, TV advertisements etc.
An analysis of what we produce should be able to tell us a lot about the way we structure our society and about our values, prejudices and so on. For example a content analysis of television advertisements of the 1970s would probably paint a far more sexist view of the World than that present today, certainly in the UK at least.
Content analysis, though it often analyses written words, is a quantitative method, it produces numbers and percentages. After doing a content analysis, you might make a statement such as "27% of programs on HFM in May 2011 mentioned at Lady GaGa compared to only 8% in 2009.”
Though it may seem crude and simplistic to make such statements, the counting serves two purposes:
· to make analysis more objective
· to simplify the detection of trends.
What constitutes content?
All content is something that people have created. We call these artefacts. You can’t do content analysis of the weather - but if somebody writes a report predicting the weather, you can do a content analysis of that. Examples could include:
Newspaper items, magazine articles, books, catalogues, web pages, advertisements, graffiti, radio programs, news items, TV programs, photos, videos, films, music, speeches, interviews, plays, gestures, products in shops and as we saw in one question, drawings
Categories and Themes
Crucial to the whole process is categorising the data. For example if you were looking at paintings produced by patients with schizophrenia you may want to categorise in terms of colour, vividness, intensity, content etc. It is possible that you will have prior expectations and might opt for a top-down approach, having decided in advance what categories to look for. More likely you would use a bottom-up method in which your categories emerge from the material produced
The Process
The sample will be the artefacts that are to be analysed. This needs to be representative. For example, if looking at gender stereotypes in car adverts that appear in magazines, you would need a wide range of different magazines to get a representative sample. If you only used men’s magazines, your sample would be biased, and you may not be able to generalise your results.
Coding System: Similarly to an observation, the researcher has to create a coding system, which breaks down the information into categories. So for the example above, for each advert, you may first identify the gender of person in the car advert, and then tally what they are depicted doing. These behavioural categories might be:
Driving the car ·
- Passenger in the car
- Washing the car
- Loading up the boot
- Sitting on the bonnet
- Looking at the car
The researcher would then tally each time either a man or woman was seen doing a particular behaviour in the advert. This is called a quantitative analysis.
An alternative to having a coding system like above is to do a qualitative analysis. This is where the researcher has categories and chooses a particular example to illustrate this category. So for the category “Driving the car” he might choose the picture above left as a demonstration. Instead of counting the data, it is described (hence qualitative rather than quantitative).
Results: the researcher then looks at the data he has collected, and draws conclusions. For example, he may find that many more men are depicted driving the cars than women, and women are more likely to be seen as passengers. He might conclude from this that there is a gender bias in the way cars are advertised.
Strengths and weaknesses of content analysis
- Content analyses tend to have high ecological validity because it is based on observations of what people actually do; real communications that are current and relevant such as recent newspapers or children’s books. Therefore, it also has high mundane realism.
- Also, as the artefacts that are being analyzed already exist, there is no chance of demand characteristics. The person who created the artefact did not know that what they created would be used in a content analysis, and therefore, this could not have affected them.
- Unlike other methods of observation, content analysis can be replicated by others. So long as the artefacts that are being analysed are available for others (the same magazines, TV shows etc), the analysis could be repeated and reliability assessed.
- Reliability can also be assessed using inter-observer reliability (see below)
- A big weakness in a content analysis, as in all observations, is observer bias. This can affect both the objectivity and validity of findings as different observers might interpret the meanings of the categories in the coding system differently.
- There can be a big culture bias as the interpretation of verbal or written content will be affected by the language and culture of the observer and the coding system used.
- Similarly to other non-experimental methods, we cannot draw cause and effect relationships
An example
Manstead and McCulloch (1981) watched 170 television advertisements in a week and scored them on a whole range of factors such as gender of product user, gender of person in authority, gender of person providing the technical information about the product and so on.
Dealing with validity in content analyses
As in other research methods, validity can be affected by having an unrepresentative sample. Remember, in a content analysis, the sample is not a group of people, but the artefacts that you decide to analyse.
To deal with this, the researcher needs to ensure that the sample is representative. For example, if looking for racial stereotypes in TV adverts, you would need a range of adverts, for a range of different products, which are shown at different times of the day and on different channels.
The coding system could also be an issue with regards to validity. The coding system may not actually be measuring what you intend to measure. We can assess how valid our measuring system by:
Content validity: are we measuring what we intent to measure? We could ask a panel of experts of assess our coding system.
In a content analysis, there may be an observer bias. The observer doing the analysis may have an idea of what they hope to find, and so have a bias to only record those bits of information that fit their theory.
One way to overcome this is to use a double blind technique, where the person doing the observing does not know that aim of the study, and so will not have a bias.
Dealing with reliability in content analyses
The reliability of a content analysis refers to how consistent we would expect the results to be. In other words, the same results should be gained if repeated. The coding system should therefore be clear and easy to use. It should be objective so that a particular behaviour will only be recorded in one category. There are a number of ways to assess reliability in a content analysis:
Test-retest: another researcher (or even the same researcher) can repeat the analysis using the same coding system and the same artefacts. If the analysis is reliable, the same results should be gained.
Inter-observer reliability: Two or more observers can analyse the same artefacts. Their results are correlated. If the coding system is reliable, there should be more than 80% agreement.
Reliability can also be increased by training the observers in the use of the coding system through practice.
Research Design and Implementation
Aims and Hypotheses
Aims
When carrying out a piece of research it is essential that you have an aim in mind. This needs to be reasonably precise, for example ‘I’m gonna study memory’ would not be sufficiently precise. However the aims are broader, or less precise than the hypotheses. A suitable aim for memory might be ‘to see if age affects the duration of STM.’
Miller’s aim was to discover the capacity of STM
Hypotheses
Psychologists start with a theory which is a general idea about a behaviour and then develop a hypothesis which makes the theory testable. Hypotheses are more precise and should be operationalised, i.e. give some clue as to how the research will be carried out.
Experimental or alternative hypothesis
This makes your prediction, for example:
‘As age increases the duration of STM decreases.’
Directional or non-directional?
Having decided on your hypothesis and aims you need to decide on the direction. In the examples above I’ve already done this. When we say that we expect ‘duration of STM to decrease as age increases’ we are making a definite prediction. That prediction has direction. Compare this to the statement that ‘duration of STM will be affected by an increase in age.’ Will duration increase or decrease? The hypothesis doesn’t say. It could go either way.
Directional
If the hypothesis has a direction we say it is ‘directional’ or one-tailed. In the first example we are saying that duration of STM will decrease.
Non-directional
If we are not prepared to commit ourselves and simply say there will be an affect then this is non directional or two tailed.
Exam tips
If you’re asked to construct a hypothesis try to be specific. Consider the IV and DV or the covariables (in the case of a correlational analysis). If the DV is memory indicate how ‘memory’ is being assessed. For example ‘alcohol will decrease the number of words correctly recalled.’ This is essential if the question asks you to operationalise your hypothesis.
If you’re asked to justify why a directional or non-directional hypothesis has been chosen, the answer should be simple. Look at the stimulus material. If no prior research is mentioned then opt for two-tailed. The researchers are unlikely to be sure about what will happen so they’re going to edge their bets. Previous research that conclusively points to a certain expectation and you’ll opt for direction (in the direction of those previous findings). Justify your choice in terms of previous research (or lack of).
For example:
Psychological research has indicated that there is a gender difference in the way individuals behave at controlled pedestrian crossings (where the crossing is controlled by a red and green light). It has been found that females are more likely than males to comply with the lights.
If asked to produce a hypothesis and justify your choice of directional or non-directional;
Males are more likely to display independent behaviour at a controlled pedestrian crossing than females.
Directional has been chosen since previous research suggests women are more likely to comply at pedestrian crossings.
If the study is correlational then use the word correlation in your hypothesis.
If you are opting to go non-directional then ‘correlation’ is sufficient. If you’re opting for directional then you will need to precede with ‘positive’ or ‘negative.’
There will be a negative correlation between age and duration of STM as determined by the B-P technique.
Selecting your victims (sorry participants)
When conducting research psychologists need participants. In an ideal world, a study would include all members of a target population as this would provide the most accurate results. A target population is a group of people who share the same characteristics eg. married women, A Level students, males over the age of 40 who enjoy playing golf.
Having decided on your method (experiment, correlation etc.) and your design (repeated or individual) you now need to decide how you will choose the people who will be assigned to your conditions or groups.
Clearly it is impossible to include all members of the target population within a study so a section of that population, a sample is included instead. If a sample is truly representative, then psychologists should be able to generalise the conclusions of the study to the whole target population. There are several ways of obtaining a sample explain how you would obtain the following samples, say why you may choose the sampling method (i.e. what is it good for?) and :
Random Sample
It is practically impossible to get a truly random sample. In a random sample every member of your target population would have an equal chance of being selected. So for example if you wanted a random sample of primary school children in the Market Harborough area you would need to obtain all of their names, put them in a hat and draw your sample out. In actual fact that would be the easy bit. The difficult task would be finding them and persuading their parents to let you chosen ones take part!
The main disadvantages of this method are:
Systematic sample (similar to random, with the same disadvantages)
This could be done by visiting the target schools and selecting every 5th child in the register. This would still be time consuming. If your target was people in MH, you could select every 20th street and then visit every 10th house in those streets etc. However, it cannot be claimed that every person in MH has an equal chance of being selected!
Stratified sample
Here each variable affecting the outcome of the procedure needs to be considered. For example if you were investigating voting intentions you would want to select on grounds of: gender, occupation, age, education, home ownership etc. So because the male:female ratio is about 50:50 your sample would be 50:50. Because about 65% of the adult population are home owners then 65% of your sample would be too. Etc., etc.
Disadvantages
Opportunity sample
Now we’ve hit rock bottom! This is probably the least effective way since it involves selecting whoever happens to be available and willing to take part!
Next year in your search for victims, chances are you’ll go to the sixth form centre and pick out a few friends or non-threatening strangers. Valentine (1982) estimates that 75% of all American and British psychology research is conducted on students, and the majority of these will have been selected in this way!
Disadvantages
Sample Size
How many people are going to be part of your opportunity sample? Large samples can be expensive and are definitely time consuming Small samples make it difficult to get a significant result The larger the sample the more likely it is that the conclusions of the investigation will reflect the behaviour of the whole target population. The size of a sample will be dictated partly by time and financial constraints, although statistical tables should be consulted to note an acceptable number of Ps in order to achieve valid, successful results.
When conducting research psychologists need participants. In an ideal world, a study would include all members of a target population as this would provide the most accurate results. A target population is a group of people who share the same characteristics eg. married women, A Level students, males over the age of 40 who enjoy playing golf.
Having decided on your method (experiment, correlation etc.) and your design (repeated or individual) you now need to decide how you will choose the people who will be assigned to your conditions or groups.
Clearly it is impossible to include all members of the target population within a study so a section of that population, a sample is included instead. If a sample is truly representative, then psychologists should be able to generalise the conclusions of the study to the whole target population. There are several ways of obtaining a sample explain how you would obtain the following samples, say why you may choose the sampling method (i.e. what is it good for?) and :
Random Sample
It is practically impossible to get a truly random sample. In a random sample every member of your target population would have an equal chance of being selected. So for example if you wanted a random sample of primary school children in the Market Harborough area you would need to obtain all of their names, put them in a hat and draw your sample out. In actual fact that would be the easy bit. The difficult task would be finding them and persuading their parents to let you chosen ones take part!
The main disadvantages of this method are:
- Time consuming
- Might not be representative
Systematic sample (similar to random, with the same disadvantages)
This could be done by visiting the target schools and selecting every 5th child in the register. This would still be time consuming. If your target was people in MH, you could select every 20th street and then visit every 10th house in those streets etc. However, it cannot be claimed that every person in MH has an equal chance of being selected!
Stratified sample
Here each variable affecting the outcome of the procedure needs to be considered. For example if you were investigating voting intentions you would want to select on grounds of: gender, occupation, age, education, home ownership etc. So because the male:female ratio is about 50:50 your sample would be 50:50. Because about 65% of the adult population are home owners then 65% of your sample would be too. Etc., etc.
Disadvantages
- Time consuming
- Not a truly representative sample
Opportunity sample
Now we’ve hit rock bottom! This is probably the least effective way since it involves selecting whoever happens to be available and willing to take part!
Next year in your search for victims, chances are you’ll go to the sixth form centre and pick out a few friends or non-threatening strangers. Valentine (1982) estimates that 75% of all American and British psychology research is conducted on students, and the majority of these will have been selected in this way!
Disadvantages
- A very poor representative or cross-sectional sample!
Sample Size
How many people are going to be part of your opportunity sample? Large samples can be expensive and are definitely time consuming Small samples make it difficult to get a significant result The larger the sample the more likely it is that the conclusions of the investigation will reflect the behaviour of the whole target population. The size of a sample will be dictated partly by time and financial constraints, although statistical tables should be consulted to note an acceptable number of Ps in order to achieve valid, successful results.
Reliability and Validity
Both of these have been mentioned during the year, particularly ‘validity’ as in ‘ecological validity’ or ‘experimental validity.’ However, you now need to fully understand what both of them mean, how they can be increased and most importantly how to remember which is which!
Reliability
Reliability is akin to consistency
If you use a meter rule to measure the length of your classroom today, and you repeat the procedure next week, you will expect to get the same result. The meter rule is consistent in its measurement or we say it is reliable!
Reliability in Psychology
This can be measured in a number of ways depending upon the circumstances. However, each time we are looking for consistency of measurement:
Reliability of observations
This year some of the students have observed aggressive acts in men’s and women’s football to see if the men’s game is really more aggressive. (Personally I never realised that real men played football but that’s a different issue).
Reliability of tests
If you measure someone’s IQ today you would expect to get a similar result if you used the same test to assess the same person in a few weeks time. If the results were the same time (i.e. if the results were consistent (that word again)), you could assume the test was reliable!
Both of these have been mentioned during the year, particularly ‘validity’ as in ‘ecological validity’ or ‘experimental validity.’ However, you now need to fully understand what both of them mean, how they can be increased and most importantly how to remember which is which!
Reliability
Reliability is akin to consistency
If you use a meter rule to measure the length of your classroom today, and you repeat the procedure next week, you will expect to get the same result. The meter rule is consistent in its measurement or we say it is reliable!
Reliability in Psychology
This can be measured in a number of ways depending upon the circumstances. However, each time we are looking for consistency of measurement:
Reliability of observations
This year some of the students have observed aggressive acts in men’s and women’s football to see if the men’s game is really more aggressive. (Personally I never realised that real men played football but that’s a different issue).
- Inter-rater reliability
- One way of tackling this problem would be for one person to watch a game played by each gender, look for various aggressive acts and score them accordingly. However, you only have one person’s opinion. Better would be to get two or three people to do it independently and compare scores afterwards. To ensure that results were reliable the raters would sit down beforehand and decide on the criteria to use and how to apply these. For example decide exactly what was meant by ‘dirty tackle’ (no jokes please) or an ‘aggressive act.’ This would ensure inter-rater reliability. Or in English it would ensure consistency in measurement between the observers. All singing from the same hymn sheet in politico-speak.
Reliability of tests
If you measure someone’s IQ today you would expect to get a similar result if you used the same test to assess the same person in a few weeks time. If the results were the same time (i.e. if the results were consistent (that word again)), you could assume the test was reliable!
- Split test reliability
- Rather than waiting a few weeks to try the test again it is possible to use split test reliability. For example with an IQ test, split it in half give both halves to the participant and compare their score on each separate half. If scores on each half are similar psychologists assume the test to be reliable.
Validity
Does the test or the experiment measure what it’s s’pose to be measuring?
We have mentioned this word ‘validity’ on a number of occasions, usually in relation to ‘ecological validity.’ However, there are a number of different types of validity; here we’ll concentrate on ‘internal’ and ‘external’, sometimes referred to as ‘ecological.’
Internal (or experimental) Validity
Are the effects that have been caused actually due to the independent variable? For example if we’ve found that coffee (the I.V.) does increase speed of reaction (the D.V.), can we be certain that this increase is really due to the coffee or could it be due to a confounding variable such as the time of day or just faster reactions of the second group etc.
Demand characteristics also create issues for internal validity. A perfect example being the Hawthorn Electrical Company when researchers increased productivity not by adjusting lighting as they thought, but simply by observing the workforce.
External validity (including ecological validity)
How much can the results obtained tell us about real life, or put another way; can we generalise our findings to the real world?
Coolican (1994) points out 4 major issues:
Clearly it is useful for a psychologist to have some idea of whether or not tests are valid. There are a number of ways this can be done:
Meta analyses: data can be collected form lots of different studies in different parts of the World and see if results are similar. For example Bouchard & McGue compared findings for IQ tests between MZ twins and found similar levels of correlation between them all.
Concurrent validity: if we are measuring IQ we could compare the scores obtained to school tests in maths and English, or we could compare the results of personality tests with assessments by a person’s friends and family.
Predictive: a test should be able to predict later performance, behaviour or personality. So again, a high score on an IQ test should be able to predict later success at school etc.
Does the test or the experiment measure what it’s s’pose to be measuring?
We have mentioned this word ‘validity’ on a number of occasions, usually in relation to ‘ecological validity.’ However, there are a number of different types of validity; here we’ll concentrate on ‘internal’ and ‘external’, sometimes referred to as ‘ecological.’
Internal (or experimental) Validity
Are the effects that have been caused actually due to the independent variable? For example if we’ve found that coffee (the I.V.) does increase speed of reaction (the D.V.), can we be certain that this increase is really due to the coffee or could it be due to a confounding variable such as the time of day or just faster reactions of the second group etc.
Demand characteristics also create issues for internal validity. A perfect example being the Hawthorn Electrical Company when researchers increased productivity not by adjusting lighting as they thought, but simply by observing the workforce.
External validity (including ecological validity)
How much can the results obtained tell us about real life, or put another way; can we generalise our findings to the real world?
Coolican (1994) points out 4 major issues:
- Population: Can we generalise from our small sample, probably all students, to the population as a whole?
- Location, location, location (Coolican only said it once): Can the results that we’ve obtained in a laboratory setting really tell us how people will behave in real life. Think back to memory experiments most of which were carried out in laboratories, or to ‘Stan the Man’ Milgram’s experiment in the labs of Yale University. Would people really behave this way in real life? (This is ecological validity)
- Measures: If we use the Eysenck Personality Questionnaire (EPQ) and measure a person as very extrovert and slightly neurotic, can we be sure that they are really like this in real life or in social situations? Similarly when we measure IQ, is the test we are using telling us anything real about the person?
- Times: Can experiments carried out 40 0r 50 years ago such as Asch, Milgram etc. still tell us anything about people today. I have mentioned how for example conformity changes over time. Wars, for example tend to bring populations together and make us more conformist as was measured following the Falklands Conflict of 1982.
Clearly it is useful for a psychologist to have some idea of whether or not tests are valid. There are a number of ways this can be done:
Meta analyses: data can be collected form lots of different studies in different parts of the World and see if results are similar. For example Bouchard & McGue compared findings for IQ tests between MZ twins and found similar levels of correlation between them all.
Concurrent validity: if we are measuring IQ we could compare the scores obtained to school tests in maths and English, or we could compare the results of personality tests with assessments by a person’s friends and family.
Predictive: a test should be able to predict later performance, behaviour or personality. So again, a high score on an IQ test should be able to predict later success at school etc.
Relationship between researcher and participant
As we’ve already seen this can cause problems, particularly in the experimental method. In the Milgram evaluation I touched on demand characteristics, the idea that simply because the participant was taking part in an experiment that this would affect his behaviour (all Milgram’s participants were ‘he’s).
Possible effects:
Participant reactivity
Put simply, participants will behave differently or unnaturally because they know they are being watched. This doesn’t just apply under experimental conditions but in any walk of life! The classic example which is well worth a read, but not necessary for the exam, is to be found on page 272 and is known as the ‘Hawthorn Effect.’
Demand characteristics
The idea that participants will behave the way they believe you want tem to behave. It could be that participants guess what the experiment is about, or at least think they’ve guessed, and this will influence their behaviour accordingly.
This was a criticism of the Milgram procedure. In Asch’s study on conformity, some of the participants said afterwards that they conformed because they didn’t want to mess up the experiment!
Orne (1962) persuaded participants to do strange, if not very foolish things. This argument is often used in the debate over hypnosis. Orne, for example, persuaded his participants to put their hands into a tank containing a supposedly very venomous snake. His most famous ‘experiment’ was to persuade participants to spend hours adding up random numbers and then getting them to tear up all their hard work!
Sometimes, of course, the reverse may be true, and for whatever reason, e.g. having been conned in previous studies, participants may deliberately seek to mess up your experiment by behaving counter to how they think you want them to behave.
As we’ve already seen this can cause problems, particularly in the experimental method. In the Milgram evaluation I touched on demand characteristics, the idea that simply because the participant was taking part in an experiment that this would affect his behaviour (all Milgram’s participants were ‘he’s).
Possible effects:
Participant reactivity
Put simply, participants will behave differently or unnaturally because they know they are being watched. This doesn’t just apply under experimental conditions but in any walk of life! The classic example which is well worth a read, but not necessary for the exam, is to be found on page 272 and is known as the ‘Hawthorn Effect.’
Demand characteristics
The idea that participants will behave the way they believe you want tem to behave. It could be that participants guess what the experiment is about, or at least think they’ve guessed, and this will influence their behaviour accordingly.
This was a criticism of the Milgram procedure. In Asch’s study on conformity, some of the participants said afterwards that they conformed because they didn’t want to mess up the experiment!
Orne (1962) persuaded participants to do strange, if not very foolish things. This argument is often used in the debate over hypnosis. Orne, for example, persuaded his participants to put their hands into a tank containing a supposedly very venomous snake. His most famous ‘experiment’ was to persuade participants to spend hours adding up random numbers and then getting them to tear up all their hard work!
Sometimes, of course, the reverse may be true, and for whatever reason, e.g. having been conned in previous studies, participants may deliberately seek to mess up your experiment by behaving counter to how they think you want them to behave.
Reducing demand characteristics
The most common ploy is called the single blind technique in which participants are not told details of the study or in which they are led to believe it’s about something different. This is a ploy I used in my research on hypnosis. (Will bore you with the details sometime soon!).
Clearly this raises ethical issues such as deception and informed consent.
Investigator effects
A confession. As a sixth former, many years ago, I spent, what seemed at the time, about five years doing titrations in A-level chemistry. Not one of the results I obtained was genuine. We calculated the ‘right answer’ for all of them and then obtained a reading that was close to this. This was cheating, and we know that results in Psychology have also been fiddled, some on a grand scale. Obtaining ‘expected results’ like this can be deliberate. Or it can happen without intent. We often find what we are expecting or hoping to find. Having decided that women are worse drivers we notice bad driving by women whilst ignoring similar driving by men. This happens in research and is called experimenter expectancy. The classic example is Rosenthal and Lawson (1964). They gave rats to students, telling some that their rats were ‘maze bright’ and could navigate a maze very quickly, and telling others that their rats were ‘maze dull’ and not very good at navigating a maze. In fact the rats were all similar and allocated to each group of students randomly.
From what I’ve said, you can probably guess the findings: Students with the supposedly maze bright rats found that their rats could navigate mazes significantly faster!
Reducing experimenter effects
The most common ploy is called the double blind technique, in which neither the participants nor the researchers dealing with the participants know the conditions etc. Obviously someone distant from the procedure still needs to know which participants are in which condition so that results can be analysed! This procedure is commonly used in drug testing when genuine medicines are compared to placebos.
Remember that on top of this there is the 'data analysis' section, which involves some number crunching. See contents page for details. However, the worksheets provided in lessons should cover this are in sufficient detail so I'll skip notes on this and concentrate on conformity instead. Hope this has not been too heavy a read, I've tried to keep it brief and to the point!
The most common ploy is called the single blind technique in which participants are not told details of the study or in which they are led to believe it’s about something different. This is a ploy I used in my research on hypnosis. (Will bore you with the details sometime soon!).
Clearly this raises ethical issues such as deception and informed consent.
Investigator effects
A confession. As a sixth former, many years ago, I spent, what seemed at the time, about five years doing titrations in A-level chemistry. Not one of the results I obtained was genuine. We calculated the ‘right answer’ for all of them and then obtained a reading that was close to this. This was cheating, and we know that results in Psychology have also been fiddled, some on a grand scale. Obtaining ‘expected results’ like this can be deliberate. Or it can happen without intent. We often find what we are expecting or hoping to find. Having decided that women are worse drivers we notice bad driving by women whilst ignoring similar driving by men. This happens in research and is called experimenter expectancy. The classic example is Rosenthal and Lawson (1964). They gave rats to students, telling some that their rats were ‘maze bright’ and could navigate a maze very quickly, and telling others that their rats were ‘maze dull’ and not very good at navigating a maze. In fact the rats were all similar and allocated to each group of students randomly.
From what I’ve said, you can probably guess the findings: Students with the supposedly maze bright rats found that their rats could navigate mazes significantly faster!
Reducing experimenter effects
The most common ploy is called the double blind technique, in which neither the participants nor the researchers dealing with the participants know the conditions etc. Obviously someone distant from the procedure still needs to know which participants are in which condition so that results can be analysed! This procedure is commonly used in drug testing when genuine medicines are compared to placebos.
Remember that on top of this there is the 'data analysis' section, which involves some number crunching. See contents page for details. However, the worksheets provided in lessons should cover this are in sufficient detail so I'll skip notes on this and concentrate on conformity instead. Hope this has not been too heavy a read, I've tried to keep it brief and to the point!
Data Analysis
We shall now consider what to do with all the data collected by various methods.
Broadly speaking we can attempt a qualitative or a quantitative analysis so I’ll kick off by considering each of these in turn.
Quantitative analysis
As the name suggests considers quantities. Some forms of research method, most notably laboratory experiments and correlations collect numbers. Times taken to complete a task, score on an IQ test, attractiveness rating of an individual, the number of people who prefer Team Jacob or Team Edward etc…
Collecting numbers usually allows for quick comparisons between groups or individuals because we can calculate means and ranges, enter numbers into tables and display our data in graphs or charts. However, although numbers allow us to see the bigger picture, 65% of the population are prepared to deliver potentially fatal electric shocks to a stooge, the often fail to provide detail.
We shall now consider what to do with all the data collected by various methods.
Broadly speaking we can attempt a qualitative or a quantitative analysis so I’ll kick off by considering each of these in turn.
Quantitative analysis
As the name suggests considers quantities. Some forms of research method, most notably laboratory experiments and correlations collect numbers. Times taken to complete a task, score on an IQ test, attractiveness rating of an individual, the number of people who prefer Team Jacob or Team Edward etc…
Collecting numbers usually allows for quick comparisons between groups or individuals because we can calculate means and ranges, enter numbers into tables and display our data in graphs or charts. However, although numbers allow us to see the bigger picture, 65% of the population are prepared to deliver potentially fatal electric shocks to a stooge, the often fail to provide detail.
Simply reporting the memory span of Clive Wearing doesn’t really give us a feel for his predicament or that of his wife Deborah. Quantitative data can be seen as cold and overly clinical, failing to give us depth of insight into our area of study. Qualitative data on the other hand provides the insight into the emotional and experiential. |
Qualitative analysis
Fills in gap left by numbers. This form of analysis should provide depth of appreciation, a better feel for what a condition, behaviour or situation entails. Typically qualitative data is collected during case studies, naturalistic observations, content analyses, surveys and questionnaires. Descriptions of cooperative behaviour in childhood play, self reports of domestic abuse and recordings of gender-typical behaviour in TV advertising would be examples.
This type of data can be difficult and time consuming to analyse. No quick-fix calculation of a median, rather looking for trends and categorisation, followed by sub-categorisation and sifting through hours of tape recordings for examples to support your categories.
We shall consider both in more detail as we look at examples. It is also worth pointing out that although certain methods generally churn out certain types of data, lab experiments: quantitative, this is not always the case and there is often overlap. Observations will often generate both, as will case studies and content analyses.
Quantitative techniques
First it is useful to make a distinction between levels of data. You’ve probably never given this much thought before, but numbers can be used in different ways to tell us different things.
Levels of data
The board don’t make it clear whether or not this is needed at AS, although it certainly is when we return to research methods at A2. However, it’s not difficult stuff and it does help when we look at graphs and charts later, so bear with me and read on…
Nominal data
This is the lowest level of data collection. Nominal simply tells us the number of people or things falling into a certain category. For example, the number of AS students doing English, the number of nine year olds able to perform liquid conservation and so on.
Nominal data results from a tally or headcount. We present nominal data in a bar chart.
Ordinal data
As the name suggests, allows us to put things in order. Consider the men’s 100 m at Beijing. The finishing order was:
First: Bolt, Second: Bailey, Third: de Lima
This is ordinal. From the data provided we can place the three runners in order… we know that Bolt ran fastest and Bailey second fastest. However, what we can’t tell from the data provided is the distance between the runners. The interval (or gap) between first and second might be very close with de Lima a distant third. We can’t tell!
Ordinal data allows us to place things in order but without the confidence that our measuring scale is accurate. Imagine placing the following personalities in descending order of attractiveness: 1. Kylie Minogue, 2. Danii Minogue, 3 Susan Boyle
From this we can tell that Kylie is tops, Danii is second, SuBo is third. What we can’t tell is the difference between them. We can’t tell if SuBo is a close third or distant or if Danii is closer to Kylie or Subo.
Just in case you’re in any doubt:
Fills in gap left by numbers. This form of analysis should provide depth of appreciation, a better feel for what a condition, behaviour or situation entails. Typically qualitative data is collected during case studies, naturalistic observations, content analyses, surveys and questionnaires. Descriptions of cooperative behaviour in childhood play, self reports of domestic abuse and recordings of gender-typical behaviour in TV advertising would be examples.
This type of data can be difficult and time consuming to analyse. No quick-fix calculation of a median, rather looking for trends and categorisation, followed by sub-categorisation and sifting through hours of tape recordings for examples to support your categories.
We shall consider both in more detail as we look at examples. It is also worth pointing out that although certain methods generally churn out certain types of data, lab experiments: quantitative, this is not always the case and there is often overlap. Observations will often generate both, as will case studies and content analyses.
Quantitative techniques
First it is useful to make a distinction between levels of data. You’ve probably never given this much thought before, but numbers can be used in different ways to tell us different things.
Levels of data
The board don’t make it clear whether or not this is needed at AS, although it certainly is when we return to research methods at A2. However, it’s not difficult stuff and it does help when we look at graphs and charts later, so bear with me and read on…
Nominal data
This is the lowest level of data collection. Nominal simply tells us the number of people or things falling into a certain category. For example, the number of AS students doing English, the number of nine year olds able to perform liquid conservation and so on.
Nominal data results from a tally or headcount. We present nominal data in a bar chart.
Ordinal data
As the name suggests, allows us to put things in order. Consider the men’s 100 m at Beijing. The finishing order was:
First: Bolt, Second: Bailey, Third: de Lima
This is ordinal. From the data provided we can place the three runners in order… we know that Bolt ran fastest and Bailey second fastest. However, what we can’t tell from the data provided is the distance between the runners. The interval (or gap) between first and second might be very close with de Lima a distant third. We can’t tell!
Ordinal data allows us to place things in order but without the confidence that our measuring scale is accurate. Imagine placing the following personalities in descending order of attractiveness: 1. Kylie Minogue, 2. Danii Minogue, 3 Susan Boyle
From this we can tell that Kylie is tops, Danii is second, SuBo is third. What we can’t tell is the difference between them. We can’t tell if SuBo is a close third or distant or if Danii is closer to Kylie or Subo.
Just in case you’re in any doubt:
Interval and Ratio data
Interval and ratio data allows us to put things in order, but this time we can be certain that the gaps between numbers are the same. We know that the difference between 4 seconds and five seconds is exactly the same as the difference between 10 seconds and 11 seconds. In psychology we might measure reaction times, weights of people, time taken to complete cognitive tasks, speeds of car drivers, drop in body temperature during stage 4 sleep etc.
Generally speaking if it’s interval and ratio a piece of technical equipment is needed, for example a ruler, scales, thermometer an EEG and so on.
We tend to summarise interval and ratio data with a line graph.
What’s the difference between interval and ratio data I can hear you asking. In practical terms we tend to treat them similarly. For example we can use more powerful parametric statistical tests when we collect interval or ratio data. However, there is a subtle difference.
Ratio data has a true zero! Your weight can’t drop below zero grams, you can’t remember something for minus 7 plus or minus two seconds….
As a result we can say that 10 seconds is twice as long as 5 seconds.
Interval ratio doesn’t have a true zero. Your body temperature could drop to minus 3 degrees Celcius (although its fair to say you’d probably be quite poorly by that stage). As a result we can’t say that 10 degrees Celcius is twice as hot as 5 degrees Celcius.
So to summarise:
NOIR
TASK: Consider the following:
Placing films in ascending order of preference, counting the number of people who prefer brown or red sauce, rating burgers on a scale of 1 to 10, recording changes in body temperature during a night’s sleep, counting the number of people who fall into the categories: ectomorph, mesomorph and endomorph, comparing the times taken for a hungry or sated rat to successfully run a maze.
Measures of central tendency
The first step in dealing with any set of numbers usually involves calculating an average. By average we usually mean mean J. Not a typo, I really did mean mean mean!
However, as you’ll all know from your GCSE numbers there are three measures of central tendency or average, (mean, median and mode) and you’ll need to know how to calculate each, and also the good and bad points relating to each. We shall consider them in turn:
Mean
The good old average. Add your numbers up and divide by the number there are.
For example, time (in seconds) taken for a rat to successfully run a maze:
23, 45, 36, 21, 23, 33, 19, 32
Added together = 232
Divided by 8 = 29
Getting a whole number was pure luck J
Good points
The mean is sensitive! Meaning it’s in touch with its feminine side… well not really. What it does mean (that word again and again) is it uses all the data. In the example above it uses all eight numbers.
However
Imagine if one of the rats had taken an eternity to run the maze… so the last one had taken 320 seconds rather than the 32. The mean now increases to 65 seconds.
Think about what we want from a measure of central tendency… a number that is central to our date perhaps? Clearly our new number (65) is not central. It is nearly twice the value of seven of our numbers.
So if we have outliers or extreme values we don’t use the mean!
The mean is most useful when the data approximates to a normal distribution:
Interval and ratio data allows us to put things in order, but this time we can be certain that the gaps between numbers are the same. We know that the difference between 4 seconds and five seconds is exactly the same as the difference between 10 seconds and 11 seconds. In psychology we might measure reaction times, weights of people, time taken to complete cognitive tasks, speeds of car drivers, drop in body temperature during stage 4 sleep etc.
Generally speaking if it’s interval and ratio a piece of technical equipment is needed, for example a ruler, scales, thermometer an EEG and so on.
We tend to summarise interval and ratio data with a line graph.
What’s the difference between interval and ratio data I can hear you asking. In practical terms we tend to treat them similarly. For example we can use more powerful parametric statistical tests when we collect interval or ratio data. However, there is a subtle difference.
Ratio data has a true zero! Your weight can’t drop below zero grams, you can’t remember something for minus 7 plus or minus two seconds….
As a result we can say that 10 seconds is twice as long as 5 seconds.
Interval ratio doesn’t have a true zero. Your body temperature could drop to minus 3 degrees Celcius (although its fair to say you’d probably be quite poorly by that stage). As a result we can’t say that 10 degrees Celcius is twice as hot as 5 degrees Celcius.
So to summarise:
- Nominal data tells us the number falling into a certain category.
- Ordinal, interval and ratio allow us to place things in some kind of order
- Interval and ratio go a step further. They allow us to put things in order and determine the difference between adjacent members on that scale.
NOIR
TASK: Consider the following:
Placing films in ascending order of preference, counting the number of people who prefer brown or red sauce, rating burgers on a scale of 1 to 10, recording changes in body temperature during a night’s sleep, counting the number of people who fall into the categories: ectomorph, mesomorph and endomorph, comparing the times taken for a hungry or sated rat to successfully run a maze.
Measures of central tendency
The first step in dealing with any set of numbers usually involves calculating an average. By average we usually mean mean J. Not a typo, I really did mean mean mean!
However, as you’ll all know from your GCSE numbers there are three measures of central tendency or average, (mean, median and mode) and you’ll need to know how to calculate each, and also the good and bad points relating to each. We shall consider them in turn:
Mean
The good old average. Add your numbers up and divide by the number there are.
For example, time (in seconds) taken for a rat to successfully run a maze:
23, 45, 36, 21, 23, 33, 19, 32
Added together = 232
Divided by 8 = 29
Getting a whole number was pure luck J
Good points
The mean is sensitive! Meaning it’s in touch with its feminine side… well not really. What it does mean (that word again and again) is it uses all the data. In the example above it uses all eight numbers.
However
Imagine if one of the rats had taken an eternity to run the maze… so the last one had taken 320 seconds rather than the 32. The mean now increases to 65 seconds.
Think about what we want from a measure of central tendency… a number that is central to our date perhaps? Clearly our new number (65) is not central. It is nearly twice the value of seven of our numbers.
So if we have outliers or extreme values we don’t use the mean!
The mean is most useful when the data approximates to a normal distribution:
Median
Place your numbers in ascending (or descending) order and pick out the middle one. How much more central could you get. If there’s an even number then find the half way point between the two middle numbers:
19, 21, 23, 24, 32, 33, 36, 45
Middle two numbers are 24 and 32. So our median is 28. If in doubt add together your two middle numbers and divide by two. 24 + 32 = 56. 56/2 = 28.
Good and bad points are the reversal of the mean:
Good points
It isn’t unduly influenced by extreme values. Substituting 32 by 320
19, 21, 23, 24, 33, 36, 45, 320.
Median now lies between 24 and 33… 28.5
The median has been adjusted by 0.5…. not gonna’ lose much sleep over that are we?
However, it ignores most of the numbers we’ve gone to all that trouble to collect. We might have hundreds of numbers and we use the middle two!
Mode
Easy-peasy lemon squeezy: which number is most common. In our first set 23 occurs twice so the mode is 23.
Place your numbers in ascending (or descending) order and pick out the middle one. How much more central could you get. If there’s an even number then find the half way point between the two middle numbers:
19, 21, 23, 24, 32, 33, 36, 45
Middle two numbers are 24 and 32. So our median is 28. If in doubt add together your two middle numbers and divide by two. 24 + 32 = 56. 56/2 = 28.
Good and bad points are the reversal of the mean:
Good points
It isn’t unduly influenced by extreme values. Substituting 32 by 320
19, 21, 23, 24, 33, 36, 45, 320.
Median now lies between 24 and 33… 28.5
The median has been adjusted by 0.5…. not gonna’ lose much sleep over that are we?
However, it ignores most of the numbers we’ve gone to all that trouble to collect. We might have hundreds of numbers and we use the middle two!
Mode
Easy-peasy lemon squeezy: which number is most common. In our first set 23 occurs twice so the mode is 23.
Sometimes: 19, 21, 21, 21, 24, 27, 33, 33, 33, 39 two numbers share the honours, three 21s and three 33s In this case we have two modes, a bimodal distribution. Often this could be the best way of describing our set of data. For example the males in some species of fish tend to be either big or small with few mediums. A bipolar distribution would best explain these. Left: a bimodal distribution |
Good points
Easy-peasy lemon squeezy! Easy to calculate.
However
Imagine the rats had taken the following times:
19, 21, 23, 23, 24, 25, 33, 33, 33
Our mode is 33! Remember we want a figure of central tendency. Too frequently the mode does not provide!
Measures of Dispersion
Tell us the spread of our data
Range
This is the simplest method of calculating dispersion. Basically it is the difference between the largest and the smallest value in the data:
19, 21, 23, 23, 24, 25, 33, 33, 33
In the example above range is 33-19 = 14
Obviously this is easy to calculate but it doesn’t consider all the data and since it is based only on the greatest and smallest values it is very, very prone to outliers!
Standard deviation
Is sooooo much better! Luckily, you do not need to know how it is calculated and will not be expected to calculate it in an examination. However, you will need to know why it’s soooo good and what it tells us!
Basically a mean is calculated and all other pieces of data are then compared to the mean. As a result it uses all of the data. After a little jiggery-pokery involving squaring and square rooting, we end up with a magical number.
To explain how good it is I shall demonstrate using the example of IQ. Tests are updated to ensure that the mean IQ is maintained at 100. The standard deviation is 15:
Easy-peasy lemon squeezy! Easy to calculate.
However
Imagine the rats had taken the following times:
19, 21, 23, 23, 24, 25, 33, 33, 33
Our mode is 33! Remember we want a figure of central tendency. Too frequently the mode does not provide!
Measures of Dispersion
Tell us the spread of our data
Range
This is the simplest method of calculating dispersion. Basically it is the difference between the largest and the smallest value in the data:
19, 21, 23, 23, 24, 25, 33, 33, 33
In the example above range is 33-19 = 14
Obviously this is easy to calculate but it doesn’t consider all the data and since it is based only on the greatest and smallest values it is very, very prone to outliers!
Standard deviation
Is sooooo much better! Luckily, you do not need to know how it is calculated and will not be expected to calculate it in an examination. However, you will need to know why it’s soooo good and what it tells us!
Basically a mean is calculated and all other pieces of data are then compared to the mean. As a result it uses all of the data. After a little jiggery-pokery involving squaring and square rooting, we end up with a magical number.
To explain how good it is I shall demonstrate using the example of IQ. Tests are updated to ensure that the mean IQ is maintained at 100. The standard deviation is 15:
If we drop one standard deviation below the mean to 85 (100-15) and move one standard deviation up to 115 (100+15), about 68% of our population will fall between these two numbers. Magically this is not just true of IQ and sds of 15! Given any set of normally distributed data, 68% will always fall within one standard deviation of the mean!!!
Standard deviation therefore doesn’t just tell us the spread it allows us to quantify that spread.
The smaller the standard deviation the closer our data is to the mean. It isn’t widely spread out, it is consistent. A large standard deviation tells us the opposite. Data is widely spread out around the mean, it is inconsistent.
If I set you a mock (let’s say out of 35) and the average mark was 24 (dreaming J) with a standard deviation of 4, that would tell me that most people (68% in fact) scored between 21 and 28. That’s a tightly packed set of data!
If the sd was 10 that would suggest there was a huge variety of scores, 68% falling between 14 and 34 and the other 32% being even more widely spread than that.
Standard deviation therefore doesn’t just tell us the spread it allows us to quantify that spread.
The smaller the standard deviation the closer our data is to the mean. It isn’t widely spread out, it is consistent. A large standard deviation tells us the opposite. Data is widely spread out around the mean, it is inconsistent.
If I set you a mock (let’s say out of 35) and the average mark was 24 (dreaming J) with a standard deviation of 4, that would tell me that most people (68% in fact) scored between 21 and 28. That’s a tightly packed set of data!
If the sd was 10 that would suggest there was a huge variety of scores, 68% falling between 14 and 34 and the other 32% being even more widely spread than that.
Showing off your quantitative data
Tables
First stage might include some organising of your raw data. You may have collected a list of times taken to complete a task or a list of numbers of words correctly recalled during a serial recall task. Usually the simplest and quickest way to do this is with a table. Often your table of raw data will be included in an appendix rather than in the main body of your report.
Graphs and charts
Bar chart
Probably the graph you ever drew at primary school. As we saw with levels of data, bar charts are the perfect way to illustrate any nominal data collected. For example, the number of year 3 children in a primary school that can correctly complete Piaget’s three mountains task. This could be compared to the number of year 5 children able to do the same.
Tables
First stage might include some organising of your raw data. You may have collected a list of times taken to complete a task or a list of numbers of words correctly recalled during a serial recall task. Usually the simplest and quickest way to do this is with a table. Often your table of raw data will be included in an appendix rather than in the main body of your report.
Graphs and charts
Bar chart
Probably the graph you ever drew at primary school. As we saw with levels of data, bar charts are the perfect way to illustrate any nominal data collected. For example, the number of year 3 children in a primary school that can correctly complete Piaget’s three mountains task. This could be compared to the number of year 5 children able to do the same.
Importantly here the y axis considers amounts or percentages whilst the x axis carries the categories. These categories can be in any order.
Line graph
Tend to be seen as more complex but this needn’t be the case. Yet again the y axis represents amounts, percentages, frequencies etc. but this time the x axis must also contain data that has a logical sequence and usually is numerical but could also be days of the week etc. that have a recognised set order. Line graphs are usually used for interval or ratio data. One of the advantages is the ability to superimpose many lines on one graph (look left). This makes comparison of different sets of data possible |
Scattergrams and correlation coefficients Correlations are best illustrated using scattergrams. The two co-variables are plotted (one across the x axis and the other up the y). Any link can immediately be seen. A perfect positive correlation has a coefficient of + 1.0, no correlation has a coefficient of 0. In the real world neither of these extremes usually exists. Coefficients are lie somewhere between 0 and +1.0 for positive correlations and between 0 and -1.0 for negative. The nearer 1.0 the higher the correlation. The board expect you to be able to guestimate a correlation coefficient. |
Curvilinear correlations
Correlations don’t always lie in a straight line. For example when we looked at the effects of anxiety on recall we saw an inverted ‘U’ shaped correlation with low levels and high levels of anxiety resulting in lower levels of recall (the Yekes-Dodson law).
Qualitative analysis
Most analysis we’ve considered this term has been quantitative. It has collected numbers in the form of number of items recalled, times taken and numbers of people suffering from various effects of privation etc. Not all information collected during research has to include numbers. Researchers may collect quotations, record interviews and get in depth material about people’s attitudes, feelings and beliefs. This section looks at how researchers deal with this kind of data.
Quantitative data (numbers) is typically collected during experiments, correlations, structured questionnaires (particularly using closed questions) and occasionally during observations and case studies.
Qualitative data (quotations etc) is typically the result of unstructured questionnaires and interviews, open questions, content analyses, case studies and naturalistic observations. More unusually it could be collected as follow-ups to experiments.
Usually researchers will record information such as interviews using either video or audio recording equipment. There may be many hours of such recordings. Collating this data is not easy.
Usually the researchers look for categories that the information can be broken into. For example if they had recorded aggressive behaviour in a primary school playground they would sift through and look for appropriate categories. These are more likely to emerge during the study than have been planned for in advance. It could be that they chose to split the behaviours into verbal and physical or male and female, provoked and unprovoked etc. Each of these could then be further subdivided. Physical: pinching, biting, pushing etc.
Evidence for each of these sub-categories could then be illustrated using quotations provided or by video evidence.
NB: quantitative data may be collected at the same time. For example the number of such incidents occurring and the age ranges involved etc.
Abrahamsson et al (2002) recorded hour long interviews with dental phobics. They analysed the data collected but with no particular hypothesis in mind. They categorised the collected data as follows:
- Threat to self respect and well-being
- Avoidance
- Readiness to act
- Ambivalence in coping
Each of these categories was then further sub-divided (a favourite ploy in analyses of this sort and in qualitative analysis in general. For example, threat to self-respect and well-being was split into threat to own health and threat to social life.
Having sub-divided they then provided real examples of each category; quotations ‘I’m worried my teeth will fall out’ would be evidence for threat to well-being.
Evaluation of qualitative data
Quantitative data generally paints a broader picture and overview whereas qualitative provides a better feel for the experience of being involved in such behaviour. Neither is better than the other, both have their place in psychological research!
Qualitative data
Strengths
Gives a good feel for the complex nature of human behaviour, emotions, disorders etc. Because people are encouraged to talk at length or are observed naturally we are given access to thoughts and behaviours that may be difficult to study in any other way. |
Weaknesses
The data may be difficult and time-consuming to analyse simply because of the huge amount of information that can be collected. Analysis may be less objective since the researchers opinions and prior-expectations, for example in choosing the categories, may drastically effect the results. In fact quantitative data can be influenced in this way too. Because of the lack of numbers, levels of statistical significance cannot be calculated. Researchers cannot be certain that their results are fluke or genuine. |
Other research methods
As already stated, many research studies use a combination of techniques. We saw this with case studies but as Cardwell and Flanagan point out, Schaffer and Emerson’s Glasgow babies study used natural observation, interviews and even occasionally experiments when mothers recorded how the children responded to a series of everyday events.
Meta analysis
The results of a number of studies (usually buy a variety of researchers) in a related area are combined to see if overall trends are visible. This can increase reliability since contradictory findings may be uncovered. However, different studies may be difficult to compare because of different sampling, design and methods used. Examples seen so far: van Ijzendoorn and Krooneberg’s research into cross-cultural variations in attachment and Deffenbacher’s research into anxiety and EWT.
Longitudinal studies
A favourite with developmental psychologists since they allow changes in participants to be measured over time. The big disadvantage is attrition. People move to different areas or become impossible to contact. Examples seen so far include Hodges and Tizard’s research into privation and Rutter’s work on the Romanian orphans.
As already stated, many research studies use a combination of techniques. We saw this with case studies but as Cardwell and Flanagan point out, Schaffer and Emerson’s Glasgow babies study used natural observation, interviews and even occasionally experiments when mothers recorded how the children responded to a series of everyday events.
Meta analysis
The results of a number of studies (usually buy a variety of researchers) in a related area are combined to see if overall trends are visible. This can increase reliability since contradictory findings may be uncovered. However, different studies may be difficult to compare because of different sampling, design and methods used. Examples seen so far: van Ijzendoorn and Krooneberg’s research into cross-cultural variations in attachment and Deffenbacher’s research into anxiety and EWT.
Longitudinal studies
A favourite with developmental psychologists since they allow changes in participants to be measured over time. The big disadvantage is attrition. People move to different areas or become impossible to contact. Examples seen so far include Hodges and Tizard’s research into privation and Rutter’s work on the Romanian orphans.