Research Methods
- Created by: Rebecca Franklin
- Created on: 29-04-14 13:42
Key Terms
Internal Validity: whether the study tested what it set out to test.
External Validity: the degree to which a research finding can be generalised to other situations.
Control: refers to the extent to which any variable is held constant or regulated by a researcher.
Mundane Realism: how the study mirrors the real world.
Extraneous Variables: if this is not controlled then the results from the study will be meaningless as these factors can effect the study.
Validity: refers to how true or legitmate something is as an explaination of behaviour.
Reliability: refers to the consistency of results
Generalisability: the point of realism when the psychological research can be used to understand behaviour in everyday life.
Participant Variable: if a variable (tiredness) affects the result of a study.
IV: this is changed or manipulated DV: factors that stay the same
Key Terms
To test reliability you ask whether the study can be replicated and whether the results are consistent.
Internal validity can be effected by extraneous variables.
External validity consists of:
- Ecological validity: whether the findings can be generalised to other settings.
- Population validity: whether the findings can be generalised to other people.
- Temporal validity: whether the findings can be generalised to other times.
Ethics
- Ethics are the science of MORALS
- They can vary between cultures
- There are fundamental ethical rules that are inplace for all cultures
Psychologists need ethics because humans and non-humans that can take part in research can often involve experiencing different emotions, therefore participants need to be protected if neccessary.
BPS (British Psychological Society) Ethical Guidelines:
These are guidelines that are set up for all psychologists at whatever level to follow.
Ethics
Pyschology Is Dead Cool Really
Protection from harm:
Psychologists have a responsibility to protect their participants from physical and emotional harm. Participants should not be exposed to more risk than they would experience in everyday life.
Informed consent:
Giving their consent means they are agreeing to something - a participant should therefore always agree to take part in a study. When the participant is told the details of the study and agrees to it, this is known as informed consent.
Deception:
A researcher should not withhold any information from the participants or mislead them inany way about the nature of the study. There are exceptions when deception is minor such as it won't show unease when the deception is revealed.
Ethics
Confidentiality:
This means keeping the information private. The participants should feel confident that the study won't reveal information about their identity. They should remain anonymous.
Right to withdraw:
Participants should be able to leave at any point during the study if they no longer want to take part. This also counts after the study, their data will then not be used in the research findings.
How do psychologists deal with ethical issues?
- For protection from harm they get a memory of the ethics committee to read over their research proposals. They also ask health questions before the study begins and offer aftercare.
- They wll ask adult participants to read and sign a consent form.
- They will debrief participants before the study begins.
- They will emphasis in the debrief the participants right to withdraw from the research.
- To ensure it is confidential, the researcher will give particpants numbers and letters in place of their name and details.
Variables
A variable is something that can change within a study.
An independent variable is what can be changed or manipulated by the researcher (cause).
A dependent variable is which is measured by the researcher (effect).
'Operationalising' variables helps other researchers to replicate your research, as you put in detail as clearly as you can how each variable is measured.
An extraneous variable is anything other than the IV that may have an influence on the DV - it is an uncontrolled variable that if possible should be eliminated. If a researcher does not control it then it may have a negative effect on the findings. Extraneous variables can include:
- Age
- Intelligence
- Gender
- Time of day
- Demand characteristics (Hawthorne effect - P's altering behaviour due to being aware)
- Investigato effects
Aims and Hypotheses
All studies have aims that show the purpose of the study, it provides clarity of what the research is trying to discover.
A hypothesis is a statement of the expected outcome of a research. There are 3 types:
- Directional Hypotheses (one tailed hypothesis):
This is when the researcher has a good idea of what is going to happen so they can predict the specific outcome. - Non-directional Hypotheses (two tailed hypothesis):
This is when the researcher is less sure of what is going to happen during the study so will predict a more general outcome. - Null Hypotheses:
This is when the researcher is sure that the independent variable will have no effect at all on the dependent variable.
Pilot Study
A pilot study is a small scale trail run of a research design/experiment before doing the real thing.
They are carried out because it allows researchers to identify any potential problems in the method chosen, instructions given, procedure, materials and measurements.
They save a lot of time and money doing this.
Pilot studies are possible to do with all research methods apart from case studies and natural experiments.This two types of research are rare so it would be to wasteful if they were to use a sample for a pilot study.
Sampling Techniques
A sample is a selection from the research population.
The research population is a group of people whose characteristics a researcher wishes to draw a conclusion from.
Sampling is the process of selecting a small selection of individuals from the research population. As the results will be generalised from this small sample, the researcher needs to be sure that their sample is representative. If the sample is not representative, it will have low population validity.
Random Sampling:
This is where everyone in the target population has an equal chance. The researcher obtains a list and then uses a computerised random generator to select the required amount.
- This reduces the chances of selecting a biased sample so improved population validity.
- It can be difficult to obtain a list of everyone and if you can not everyone may want to participate or is available.
Sampling Techniques
Opportunity Sampling:
This is where the researcher selects anyone who is available and is willing to take part. People asked are those who are most convinient for the researcher to ask (e.g. if the researcher works in a school they may ask the students).
- This is a time and cost effective technique as participants are readily available, it also means they could have a larger sample size as it is cheaper to get participants making the sample more representative.
- Using the example of a researcher using students, they may feel obliged to take part, this means that ethical issues may arise with right to withdraw and giving their consent.
Volunteer Sampling (Self-selected Sampling):
Here participants put themselves forward, maybe after seeing an advert.
- This is a convenient way to collect a sample and from a researcher it is not biased.
- A weakness is that only people who see the advertisment have the opportunity that may significantly reduce the sample size and potentially significant findings.
Experimental Methods - Laboratory
These tend to take place in a spcial facility such as a laboratory within psychology departments. The independant variable is directly manipulated and the dependant variable is directly measured. All extraneous variables are controlled as much as possible. Participants are aware they are being studied.
- Lab experimentcs cannot be used in all situations, this is due to independant variables being manipulated. This cannot be done is situations where it would be impractical or unethical meaning this method cannot be used in all research cases.
- Being able to control variables makes the research very articfical, meaning participants involved may not display natural behaviour and demand characteristics may arise therefore reducing the ecological validity of the researches findings.
- A positive to lab experiments is that they can be easily replicated, this is due to the strict control and well written procedure. This means that findings can easily be tested for reliability.
Experimental Methods - Field
This type of experiment is natural, meaning participants are unaware of them monitored.
The IV and DV can still be manipulated however it is much harder to control extraneous variables with this method.
- A strength of this method is that experiments are set in real life environments, this means that results from research have higher ecological validity. Therefore findings are able t be generalised as they relate better to the real world.
- A weakness to participants being unaware they are being monitored is that they would not have been debriefed meaning they are unaware of their rights which could cause them distress throughout the experiment. This means researchers have to be careful when carrying out field experiments and need to assess whether they could cause harm to any of their participants.
- A strength to using this method is that the chance of demand characteristics is greatly reduced. As participants are unaware they are being observed they will act more naturally which overall makes the findings of the research valid.
Experimental Methods - Natural
The researcher does not manipulate the independant variable at all so that it occurs naturally.
They only meaure the effect the IV has on the dependent variable.
- A strength to this research is that investigators are able to look at topics that may be impractical or unethical using a different experimental method. This is due to no manipulation of the IV, meaning they are able to get information on topics that may not have any research into before.
- Participants are unaware that they are taking part, this increases the ecological validity of the experiment meaning results can be generalised to other situations.
- A weakness to participants being unaware they are being observed is ethical guideilnes that allow participants to give their consent. This means the guidelines are being breached so researchers need to be very careful if they wish to go along with their experiment.
Experimental Designs
Once a researcher has decided their experimental method, they need to create a hypothesis and decide on their experimental design.
An experimental design is how to allocate participants to different experimental condidtions.
Most experiments involve 2 conditions, these are 2 versions of the IV E.G to learn a list of items you could either use words or pictures.
There are 3 types of experimental design:
- Independant groups
- Repeated measures
- Matched pairs
Experimental Designs - Independent Groups:
This is where different participants are placed in each condition.
- A weakness to this design is the number of participants required, due to the two conditions needing different participants making the design expensive. However this can be a positive as it will increase the number of participants meaning validity increases due to more participants involved in the findings.
- A weakness to this design is individual differences, as different participants are being used in each condition it could mean that participant differences will have a negative effect on results. This means conclusions of the research will also be affected making the research invalid.
- Participant variables such as tiredness and boredom are significantly reduced due to them being in only taking part in one condition. This means that the validity of the research in increased.
Experimental Designs - Repeated Measures
This is where participants take part in both conditions.
- A weakness to this design is the that participants experience both conditions so order effects such as boredome can have a negative effect on the results. This overall makes findings from the research lack validity.
- A strength to this study is that individual differences is eliminated in effecting the results. This is due to participants taking part in both conditions so results can be compared meaning the results of the stdy are not effected.
- Due to participants taking part in both conditions, it means that they do not need as many people. This makes the design cost effective as only one set of participants are required.
Experimental Designs - Matched Pairs
This is where different participants are used in each condition but they are matched to a condition for characteristics (key variables e.g. age, gender) that may have an effect on their performance.
- As participants only take part in one condition, it means that order effects such as boredom are reduced, meaning results from the research have higher validity.
- A weakness to this design is that it is time consuming. It takes time to match the participants and it can be difficult meaning it can also be inaccurate that may affect the researches findings overall having a negative effect on its validity.
- A strength to this design is due to participants only being in one condition it can make the results fairer. This is because they only need to produce one set of materials making this design more fair and less time consuming.
Questionnaires
- This is a self-report technique which allows participants to directly provide information about themselves.
- They are highly replicable as they are easy to ensure procedures are the same for all participants. This means researchers are able to check findings for reliablilty of their study.
- They are time and cost efficient due to a large number of participants being able to be reached in a short space of time. Therefore lots of data can be collected making this research method efficient.
- A problem with this research method is that samples can be biased, this reduces population validity therefore meaning samples can be unrepresentative.
For a questionnaire to be good it needs to have:
Clarity (this means the participant is able to understand the questions), analysis (means that the question easy to analyse and you won't get lots of different answers back), bias (meaning the question wont be swayed so that the participant will answer it in a certain way.
Interviews
- Self report technique where the interviewwe asks the participant questions verbally. There are structured and unstructured interviews.
- Structured interviews have pre-determinded questions and unstructured is where new questions are developed as they go along.
- This is a good method for dealing with complex and sensitive issues, this is because the researcher can tell whether a person is destressed or not meaning this method can be controlled.
- Interviews are a very time consuming way to collect information asit requires you to make the questions, ask them and then have to analyse information , however you collect rich data by using this method allowing you to have more detail on your study.
- A limitation to interviews is people petray the demand charateristics that may not be what the are like naturally, this is also refered to as 'social desirability' and its very common as they are face to face. Overall this effects the validity of the research.
Case Studies
- These are in-depth studies of one individual.
- They allow researchers to investigate unique cases.
- Researchers may use a variety of different research method techniques to find information out about the participant.
- An advantage of case studies is that they allow researchers to investigate topics that may be unethical to investigate experimentally, E.G. Genie case study. Therefore reserachers are able to investigate topics that may not be ethical to look at meaning they discover more things about human behaviour and memory.
- Due to case studies being in-depth cases of one individual, it makes it difficult for researchers at the end of the study to generalise their results. Also due to the information coming from a case study, other researchers are unable to replicate the research for reliability, meaning studies coming from case study research can never be confident generalised as an explaination of behaviour.
- Finally researchers using case studies need to especially careful with ethical issues, this is because participants are usually very young or have a severe disorder, therefore they need to ensure they follow BPS guidelines strictly to ensure there investigation is carried out with no problems.
Observations
- This is where behaviour is watched and recorded.
- They can either be naturalistic observation, where participants are in their natural environment without any manipulation or a controlled observation where it takes place where variables are controlled and manipulated by the experimenter.
- The researcher also needs to decide whether it will be unstructured (they will observe everthing that is going on) or structured (they decide in advance what behaviours they will record). Within structured sampling the experimenter decides where they will do timed sampling where they see how many times a certain behaviour is performed in a given time or event sampling where they count the number of times it happens.
- An advantage of carrying out a naturalistic observation is that participants are unaware that they are being reserved, this means they will be acting completely natural therefore increasing the researches ecological validity.
- A limitation of a controlled observation is that participants are aware they are being watched, they then may become unnatural and perform demand characteristics, this means that the research findings will lack ecological validity.
- If participants are unaware they are being observed, it means that ethical issues will not have been covered. This includes their consent, etc, meaning that the research findings may not be used if any indivdual is harmed.
Content Analysis
- This is where the researchers analyse the content of something in order to conclude a pattern or trend.
- To do this the researcher first has to decide what sample they are going to analyse and then what themes may emerge from it and create a system based on this. They then count the frequency of a theme and statistical analysis can then be carried out.
- As data is collected and analysed in the simpliest way, it allows information found to be understood easily. This allows other researchers to use the information or to replicate it for reliability.
- A limitation is that reserachers may impose their own opinions on the data collected in order to conclude it, this means that the conclusion can be inconsistent which overall lowers the validity of the findings.
- If samples collected are large, it means that this process can be very time consuming as they need to prepare and then spend time analysing the data. This is difficult due to the end result may also be biased, this suggests that this research method may not be time worthy.
Correlation Analysis
- This is a statistical technique that is used for investigating the strength of relationship between two variables.
- Their results will either show a positive correlation, a negative correlation or no correlation.
- A correlation can be shown through scatter graphs or correlation coefficients (this is a number that has a maximum number of +1 and minimum of -1 and this shows the strength of the relationship).
- An intervening variable is something that comes between the two variables which is used to explain their relationship.
- This research technique can be used when a topic is unethical or impractical to manipulate its variables, meaning it can make use of existing data without threatening the health of participants, therefore making this technique practical for investigating sensitve topics.
- A limitation to correlational studies is that it doesn't find the cause and effect, this means people may assume things about findings concluded from a research. It is not possible to find the cause and effect through this research meaning this technique is not practical if you wish to find this.
- As a technique it is easily replicated, this means that studies are able to get reliable results as many people can carry out the study which can confirm a researchers findings making their study reliable and trusted.
Sampling Techniques
Stratified Sampling:
This involves classifying the population into categories and then choosing a sample which consists of an equal number of participants from each category.
- A strength to this sampling technique is it is more representative of the wider population as all categories are represented, meaning the sample has higher generalisability than opportunity sampling.
- However this technique is difficult to do and therefore takes a lot of time. There is a chance that it may not be accurate and can be researcher bias, therefore reduces the sample generalisability.
Data Analysis
Quantitative Data:
This is information that can be analysed numerically. It is behaviours measured in numbers/ quantities (how oftern, how much, how long? etc).
- This form of data is very easy to analyse and patterns are easy to see.
- However it can oversimplifyy behaviours and can lead to a loss of rich useful data.
Qualitative Data:
This is information is in narative form. It is behaviours measured in words.
- With this data you get complex and detailed information meaning you can achieve a good overview of human behaviour.
- However as it is so detailed, this information is very difficult to analyse so patterns are difficult to see.
Analysis of Quantitative Data
Measures of Central Tendency:
These analyse how close scores are to the average participants response. The 3 types of central tendency are:
- The Mean: Add all of the values together and divide by the number of values.
This makes use of all the values in final calculations however it can be misrepresentative of the data as a whole if there are extreme values. - The Median: Here you put the values in order and find the middle score.
This is not effected by extreme scores however not all values are reflected. - The Mode: This is the most frequent score.
This is not effected by extreme scores however it doesn;t use all the data.
Analysis of Quantitative Data
Measures of dispersion:
This is how spread out the data is.
Range:
- The difference between the highest and the lowest score (subtracting from each other).
- his is easy to calculate however can be affected by extreme values.
Standard deviation:
- It measures how much data from an individual is spread around a central value (the mean). A large score tells us there was a lot of deviation and a small score says participants would have been responding in a similar way.
- This is a more precise technique to measure dispersion as all scores are taken into account however it is not quick or easy to calculate.
Presentation of Quantitative and Qualitative Data
Quantitative:
- Tables
- Graphs
Bar charts, histograms, line graphs and scatter graphs
Qualitative:
This type of data is sometimes difficult to analyse as it can be a lot of interpretation that may be biased, because of this it is often converted into quantitative data via content analysis:
By transforming data patterns are easier to see and statistical analysis can be carried out for it to be summarised easier.
However detail is lost when information is converted into numbers and care needs to be taken to avoid biased coding of catagories and sample use.
Comments
No comments have yet been made