Research methods Flashcards
What’s the difference between a lab and field experiment
Lab experiments are conducted in a controlled environment, field in a more natural setting
Which experimental design is Counterbalancing used in
Repeated Measures Design
In an exam, all IV’s and DV’s should be ____
Operationalised
What’s a Lab experiment
Defined by the high level of control the researcher has over all the variables in the study
Advantages of a Lab experiment (3)
- High internal validity because extraneous variables are controlled
- The studies easily replicable due to the use of standardised procedures
- Cause and effect relationships are easily determined due to the isolation of the variables
Disadvantages of lab experiments (4)
- Lack ecological validity
- Lack mundane realism
- Participants know they’re in a study- could cause demand characteristics
- Results may be affected by experimenter bias
Advantages of Field experiments (2)
- Participants will act naturally and be completing a more usual activity leading to a high external validity
- Less demand characteristics as the participants don’t know they’re in an experiment
Disadvantages of Field experiments (3)
- Lack control over possible extraneous variables that can impact the dependant variable
- More time consuming, expensive and difficult to replicate
- Low internal validity due to sample issues (harder to randomly assign participants to certain conditions which could lead to a change in the DV due to participant variables)
What do lab and field experiments have in common
The researcher manipulates the independent variable between conditions and measures the effect on the DV
What’s a Natural experiment
The two levels of IV’s occur naturally without the researchers influence. The researcher simply records the change in the DV.
Advantages of Natural and Quasi experiments (2)
- Only way to test certain things that are deemed as unethical
- No demand characteristics* as it’s a real behaviour in the real world (high external validity*)
- = Unless Quasi-experiment was done in a lab
Disadvantages of Natural and Quasi experiments (4)
- Extraneous variables that can’t be controlled impact the DV*
- Replication difficulties (often rare events in natural exp./ measuring natural characteristics in quasi)
- Ethical issues (informed consent)
- Sample issues may not be comparable*
- = Unless Quasi-experiment was done in a lab
What’s the difference between a natural and quasi experiment
In a natural exp, the two levels of IV occur naturally, the researcher simply records effect on the DV, in a quasi, the IV already exists in the participants they’re studying and extraneous variables CAN be controlled
What do natural and quasi experiments have in common
Independent variables can’t be manipulated- they’re natural
What’s a Quasi experiment
Participants cannot be randomly assigned between levels of IV. Often because the level of IV is an innate characteristic of participants.
Observation definition
The researcher watches and records spontaneous/natural behaviour of participants without manipulating levels of indépendant variable
What’s the difference between experimental and non experimental methods
In an experimental method, the researcher manipulates the IV
What are the types of observation
Controlled vs Naturalistic observation
Overt vs Covert observation
Participant vs non participant observation
What’s controlled observation
Aspects of the environment are controlled in an attempt to give participants the same experience. This is often conducted in a laboratory setting.
Advantages of controlled observation (2)
- Controlling the environment reduces the chance extraneous variables are responsible for observed behaviour
- Results are likely to be reliable as using the same standardised procedure
Disadvantage of controlled observation
• The artificiality of the environment may result in unnatural behaviours
What’s naturalistic observation
Takes place in the real world places participants are likely to spend their time such as school or work or even their own home
Advantages of Naturalistic observation (2)
- High realism; participants are more likely to show natural behaviour
- High external validity, behaviours more likely to be generalisable to other situations
Disadvantage of Naturalistic observation
• Uncontrolled extraneous variables may be responsible for behaviour, resulting in lower internal validity
What’s overt observation
The participants can see the researcher and are aware their behaviours being used as part of an observational study
Advantage of overt observation
• Ethical through informed consent, participants agreed to take part in research
Disadvantage of overt observation
• Demand characteristics are present as they know the researchers watching or social desirability bias may be a factor
What’s covert observation
The participants are not aware they’re being observed as they can’t see someone making notes/ recordings
Advantage of covert observation
• likely to show naturalistic behaviour free from demand characteristics and social desirability bias
Disadvantage of covert observation
• More unethical as participants cannot give consent because they don’t know they’re being observed
What’s participant observation
The researcher joins the group being observed and takes part in the groups activities and conversations
Advantage of participant observation
• By taking part, the researcher may build rapport, more trust and comfort leads to more natural behaviour and disclosing more information
Disadvantage of participant observation
•Researchers can lose objectivity. Can see only from a participant perspective (‘going native’)
What’s a non participant observation
The researchers separate from the participants recording observations without taking part in the group activities
Advantage of non participant observation
• The researchers more likely to remain objective in their interpretation of the participants behaviour
Disadvantage of non participant observation
• Due to a lack of trust/ rapport the researcher misses out on important insights and participants don’t behave naturally
What’s a self report technique
The participants reveals personal information about themselves (eg. behaviours, emotions, beliefs, attitudes and memories) in response to a series of questions
What’s an interview
Participants give information in response to direct questioning from the researcher. Can also be conducted in person over the phone/video call.
What’s a questionnaire
Participants give information in response to a set of questions that are sent to them. This can be in the post or completing a form online.
Advantages of self report methods (2)
- When the same set of questions are used they’re easy to replicate
- The use of closed questions allows data analysis and the use of closed questions give the participant the opportunity to freely report their experience
(Disadvantages of self report methods) What are the 4 types of bias self report methods are prone to
- Social desirability bias- participants responding in a way that makes them look good in front of the researcher
- Demand characteristics- it’s often easy to work out the aim from the questions, people answer what they think the researcher wants
- Researcher bias- the researcher interpreting open question responses in a way that confirms their beliefs (this can be unintentional)
- Investigator effects- the personal characteristics or the body language of the interviewer may influence participants answers
Considerations of designing a self report study (5)
- Avoiding complex terminology- participants may not understand terms used and feel too embarrassed to ask for an explanation (interview) or be unable to and therefore give inaccurate responses (questionnaire)
- Rewording questions- using a skilled interviewer means questions that aren’t understood can be reworded in a way that doesn’t change the meaning so responses can still be compared to other participants
- Leading questions- these bias the response in one direction. To avoid this questions should be written in a way that doesn’t suggest a ‘correct’ way of responding. (Eg why did you find that task difficult?/ what did you prefer about condition A?)
- Piloting questions- running a small scale version of the int/q’aire can identify questions that are confusing, give away the aim or don’t produce useful/detailed responses. These can then be changed before the larger study
- Filler questions- questions not linked to the research aim but can be added to interviews to build rapport before more challenging questions or can be used as red herrings in either to help hide the research aim reducing demand characteristics.
What’s an advantage of an interview over a questionnaire
• Can build rapport so the recipient takes it more seriously
What’s an advantage of a questionnaire over an interview
• Don’t require a trained interviewer and can be distributed easily making large data collection cheap and easy
What are the three types of measurement scales in questionnaires?
- Likert Scale- level of agreement (EG. strongly agree-> strongly disagree)
- Rating scale- strength of feeling (EG.very entertaining-> not at all entertaining)
- Fixed choice option- includes a list of possible options and respondents tick all that apply
What’s a structured interview?
The interviewer reads out a list of prepared questions as they’re written
What’s an unstructured interview?
No set list of questions, an open conversation about the topic
What’s a semi-structured interview?
Combination of prepared questions with ability to ask additional questions
Advantages of a structured interview/ Disadvantages of an unstructured interview
1) Structured- interviewer doesn’t have to be trained
Unstructured (and semi-structured)- interviewer must be highly trained to think of appropriate questions in the moment
2) Structured (and semi-structured)- interview responses are easy to compare because the same questions are used
Unstructured- every interviews different making comparisons harder
Disadvantages of structured interviews/ advantages of unstructured interviews
- Structured- responses by the participant can’t be followed up with additional questions that can provide more detail
Unstructured (and semi-structured) the interviewer can ask additional questions to interesting answers - Advantage of unstructured and semi-structured: rapports more likely, the participant therefore feels more comfortable and therefore is more likely to answer questions
What’s a correlation
A correlations a method used to analyse the association between two variables (co-variables)
What’s the difference between an experimental and correlational study
Experimental designs require the manipulation of the independent variable and a measurement of the resulting change in the dependant variable. In a correlational study, no variables are manipulated, two co variables are measured and compared to look for a relationship, no causal conclusions can be drawn
Two ways to measure correlations
Scattergrams and correlation coefficients
What’s a correlation coefficient and what are the 3 types of correlation
The measure of the extent of the correlation that exists between co variables, it has a numerical value between +1 (strong positive) and -1 (strong negative)
1. Positive correlation
2. Negative correlation
3. Zero correlation
Positives of a correlation (3)
- Justify a further research: highlight potential causal relationships these can then be tested through experimental methods to discover a cause and effect relationship
- Can be used when it’s unethical or impractical to manipulate variables
- Correlation coefficient is a useful tool in describing both the direction and strength of relationship factors
Negatives of correlation (2)
- Correlation does not show causation- we don’t know what variables impacting the other or if the relationships being impacted by a third party/ variable
- Correlating can be misused or misinterpreted. Eg. In the media relationships between variables are sometimes presented as causal facts when they aren’t (high crime rate among children from single parent families)
What’s a case study
Range of data collected from an individual, group or institution. Data is mainly collected using interviews and observations, but content analysis can be performed on written evidence and even experimental techniques can be used.
How is data mainly collected in case studies
Interviews and observations
What are case studies conducted on (4)
- Psychologically unusual individuals
- Unusual events
- Organisational practices
- Typical individual within a demographic
What types of data are found in case studies
• Usually qualitative due to the use of interviews
• Quantitative can be included with the use of experimental techniques, observations and content analysis
What are the two types of case study
• Snapshot case studies
• Longitudinal case studies
What type of case study looks at behaviour over a short period of time
Snapshot case study
What’s a snapshot case study
looks at behaviour over a short period of time
What type of case study follow participants over many years
Longitudinal case studies
What’s a longitudinal case study
follows participants over many years
What’s an advantage of longitudinal case study
Shows behaviour change over time
What’re two disadvantages of longitudinal case studies
Difficult to continue long term due to
1. Funding
2. Researcher may die of old age
What’s clinical psychology
The unusual behaviour of individuals with brain damage indicate the area that’s damaged is in some way related to that behavioural function
What’s an example of case study in clinical psychology
Tan (Louis Leborgne)
What’s an example of a car study in psychodynamic psychology
Little Hans
What’s childhood psychology
Children with an unusual upbringing can be used as evidence for theories on child development
What’s an example of a case study in childhood psychology
Genie- deprived of care until 13 years old
What are the advantages of case studies (4)
- Results of one unusual case study can upend an established theory
- Often the only way to investigate very unusual/ extreme human behaviour which can’t be replicated in a lab due to ethical reasons
- Used to develop hypotheses that can be tested experimentally (Eg. Broca’s area)
- As case studies are in depth and mostly qualitative, the range of data results in realism. This holistic approach is favoured by humanistic psychologists arguing the depth of detail gives highly valid insights and a true reflection of a persons experience
What are the disadvantages of case studies (4)
- Interviews often form a large part of case studies; social desirability bias and people relying on memories which may be inaccurate
- Findings from one unusual case study can’t be generalised, other unknown factors could have influenced behaviour
- As subjects are unique, exact replications to check for reliability are not possible
- Researcher bias; researcher decides what findings to include/ exclude from the research, potentially only including data that confirms the theory, they may also lose the ability to think objectively when working with the subject due to working closely with them for long periods of time
What’s content analysis
An indirect observational method that’s used to analyse human behaviour, investigating through studying human artefacts
What’s an artefact
Something people make
What does content analysis often involve
Written word, non numerical qualitative data or transcripts being transformed into quantitate data
How do you perform a content analysis (5 steps)
- Decide a research question
- Select a sample from a larger quantity of all possible data
- Coding- the researcher decides on objective, operationalised categories/ coding units to be recorded based on the research question
- Work through the data, read the sample and tally up the number of times the pre determined categories appear
- Data analysis can be performed on quantitative data to look for patterns
How do you test for reliability in content analysis
• Test-retest reliability or inter-rated reliability
• How closely the two sets of data match in each method is assessed with a correlation test
What correlations usually accepted as reliable data in a content analysis
0.8
Three advantages of a content analysis (and thematic analysis)
- High external validity with generalisable findings as artefacts are taken from the real world
- Artefacts taken from real world so it’s easy to gather a sample
- Other researchers should be able to replicate a content analysis using the same coding units and behavioural patterns with an easy to access sample
Two disadvantages of content analysis (and thematic analysis)
- Researcher/observer bias must read subjective text- tend to interpret in a way that supports their views
- Data may lack validity- not made for research so not created under controlled conditions
What’s a thematic analysis
A type of content analysis where researchers start by attempting to identify the deeper meaning of the text by reading it first and allowing themes to emerge
How do you perform a thematic analysis (3 steps)
- Collect text and turn recordings into text through transcription
- Read text/transcripts first to spot patterns that can be coded and collected
- Re-read the transcriptions/ codes looking for emergent themes
What’s important to note when writing thematic analysis’
You must make it clear themes are not pre-determined by the researcher but come from the text
What’s an advantage of using a thematic analysis over content analysis
Theories come after discovery of themes so it can be argued that this removes observer bias
When should a researcher only use a directional hypothesis
If there’s previous research which suggests which way the results are likely to go
How do you test hypotheses
• Data’s collected and statistical testings conducted on said data providing evidence
• If the evidence is strong enough the null hypothesis can be rejected and the alternative hypothesis is accepted
What are the 6 types of hypothesis
- Directional experimental hypothesis
- Directional alternative hypothesis
- Non directional experimental hypothesis
- Non directional alternative hypothesis
- Null experimental hypothesis
- Null alternative hypothesis
What’s the difference between an alternative and experimental hypothesis
• Experimental hypothesis states there will/won’t be a difference
• Alternative hypothesis states there will/won’t be a correlation
What’s sampling
The difference between the population and the sample
What’s a target population
Every member of the the group that the investigator plans to study as the whole target population can’t be studied
What’s generalisation of a sample
Researchers conduct their experiments on a smaller sample of patricipants taken from the larger population
How is it determined whether results of a research study can be generalised or not
Whether the samples representative of the target population meaning the sample in the study shares characteristics with members of the target population
What are the five types of sampling
- Random sampling
- Systematic sampling
- Opportunity sampling
- Volunteer sampling
- Stratified sampling
What’s random sampling
Each member of the population has a mathematically equal chance of being in the sample
Method of random sampling
- Researcher collects a full list of names of the target population
- All names are entered into a container or random number generator
- Names are drawn/generated and selected from the sample
How do you reduce (can’t eliminate) effects of chance (2)
- Use a large sample size
- Statistical analysis of data
Three tests you must be able to interpret in regards to sampling
- Chi-squared
- Correlation coefficient
- Students T-test
Advantage of random sampling
No researcher bias as they can’t choose patricipants they want
Disadvantages of random and systematic* sampling (2)
- Difficult/ time consuming to get a full list of names *(if not already a list there)
- Potentially a chance of an unrepresentative sample
What’s systematic sampling
Every ‘nth’ participant is chosen from a list of the target population
Method of systematic sampling
- Researcher needs a full list of the entire population
- The researcher reads down the list selecting every ‘nth’ participant until the samples chosen
Advantages of systematic sampling (2)
- No researcher bias- can’t choose patricipants they want
- Quick method if there’s a pre existing list
What’s opportunity sampling
The researcher directly asks available members of the target population to take part. Likely individuals the researcher has easy access to or is familiar with
Method of opportunity sampling
- Researcher directly asks members from within the target population that they have access to to take part in research
- Any individuals who agree to take oath are added to the sample until the required numbers met
Advantage of opportunity sampling
Fastest way to get a sample
Disadvantages of opportunity sampling (2)
- Researcher bias
- Unlikely to be representative
What’s volunteer sampling also known as
Self selecting sampling
What’s volunteer sampling
Patricipants offer to take part after finding out about research (they’re not directly asked)
Method of volunteer sampling
- Advertisements are placed where they’re likely to be seen by members of the target population
- The adverts will include contact details and the researcher will enrol the volunteer into the sample when contacted
Advantages of volunteer sampling (2)
- Can reach a large number of patricipants
- Easy sample to collect
Disadvantage of volunteer sampling
Volunteer bias therefore not generalisable to the general population
What’s stratified sampling
By selecting from within strata, characteristics of patricipants within the sample are in the same proportion as found in the target population
Method of stratified sampling
- Strata/subgroups are identified along with their proportion in the target population (eg. Gender/Ethnicity)
- Random samplings used to select the number of patricipants required from within each stratum
Advantages of stratified sampling (2)
- Representative sample
- Avoids researcher bias (random sampling)
Disadvantages of stratified sampling (2)
- Time consuming
- Not ever characteristic can be included- researcher decided which strata are important- bias
Why do we use sampling
To make valid generalisations about behaviour we’re studying. We use methods to minimise costs while maximising generalisability
What happens if participants refuse to be in a sample
Leaves the researcher with a biased sample of only patricipants that want to take part
What are pilot studies
Small scale practice investigations carried out prior to research
What’s the point of pilot studies
The results are irrelevant, the researchers using the study to identify problems with design, method or analysis so they can be rectified
Reasons for pilot studies (6)
- To check the IV has been manipulated correctly
- To check the best method to mesure the DVs used
- To check the test/ measure is appropriate
- To check the patricipants understood instructions
- To ask patricipants about their experience
- To avoid wasting time or money
What’s experimental design
How we use the participant sample in combination with different levels of dependent variables and how we allocate those participants to conditions in an experiment
What are the two different experimental conditions
- Experimental condition
- Control condition
What are the three experimental designs
- Independent groups design (IGD)
- Repeated measures design (RMD)
- Matched pairs design (MPD)
What’s independent groups design
• Different patricipants are used in each condition so each participant only completes one condition
• patricipants are randomly allocated to each condition to avoid researcher bias
What type of data does independent groups design produce
• Unrelated data
• The individual data points in one condition cannot be pairs with any data points in the other
Advantages of independent groups design (3)
- No order effects
- Reduced chance of demand characteristics
- Time saved by random allocation
Disadvantages of independent groups design (2)
- Need twice as many patricipants
- Participant variables, extraneous ariable if more patricipants with a certain trait are in one condition
What’s repeated measures design
The same patricipants complete all conditions, each patricipants acts as their own control and are tested against themselves
What type of data does repeated measures design produce
• Related data
• Each patricipants data point can be paired with their own data point in the other condition
How do we control for order effects
Counter balancing
What’s counter balancing
Counter balancing attempts to control for (does not eliminate) order effects. It uses an ABBA format, half the patricipants complete condition A then B and the other half B then A
Advantages of repeated measures design (2)
- Twice as much data- each participant produces 2 sets of data
- No group differences
Disadvantages of repeated measures design (2)
- Order effects- first condition influences second condition
- Increased chance of demand characteristics, more likely to work out aim
What’s matched pairs design
• Different patricipants complete the conditions so each participant completes only one condition
• Patricipants are first ranked on a characteristic (eg. Aggression) and then the top two (and each of the following two) are randomly assigned to separate conditions
What type of data does matched pairs design produce
• Related data
• Each patricipants data point in one condition can be paired with the data point of the participant matched to them in the other condition
Advantages of matched pairs design (3)
- No order effects
- Reduced chance of demand characteristics
- Reduced participant variables
Disadvantages of matched pairs design (3)
- Still some participant variables
- Time consuming
- Twice as many patricipants as RMD
Whats a positive correlation
When the two variables increase or decrease together, as one increases/decreases so does the other
What’s a negative correlation
As one variable increases the other decreases
What’s a zero correlation
No relationship between the two variables
Three types of correlation
- Positive correlation
- Negative correlation
- Zero correlation
What are the three levels of measurement
- Nominal
- Ordinal
- Interval and ratio
What’s the least precise level of measurement
Nominal
What’s the most precise level of measurement
Interval (and ratio)
What’s nominal data
The number of items in each category:
The frequency count of a particular variable
Characteristics of in nominal data (3)
- Discrete variables (don’t overlap)
- Categories have no natural order
- Can’t discuss differences between each category
Examples of nominal data (3)
- Country of birth
- Career choice
- Music taste
What’s ordinal data
It has the same properties as nominal data (also a form of categorical data) however the categories have a natural order
Characteristics of ordinal data (2)
- Categories have a natural order
- The difference between each point on an ordinal scale is not consistent
Examples of ordinal data (3)
- Positions in a competition
- Choices on a Likerr scale
- Rating height among a group
What’s interval data
Interval scales are precise due to having equal interval between each adjacent point on the scale used and is not limited to a small set of discrete categories
What’s ratio data
Interval data with an absolute 0 point
Which levels of data can you convert between
A higher level to a lower level of measurement
(Interval> Ordinal)
(Ordinal> Nominal)
Who publishes ethical guidelines in the Uk
British psychological society (BPS)
Who publishes ethical guidelines in the US
American psychological association (APA)
What’s the BPS code of ethics designed to do
‘Designed to inform and assist our members in the practical and professional application of psychology’
What are the 6 ethical issues
- Informed consent
- Deception
- Right to withdraw
- Protection from harm
- Confidentiality and anonymity
- Debrief
What’s informed consent
Patricipants should get sufficient details so they can make an informed decision on whether or not to participate
Who cannot give consent (3)
- Under the age of 18 (parental consent)
- Influence of drugs/ alcohol
- Deemed mentally unfit (severe mental illness)
What’s deception
Withholding information or misleading patricipants is unacceptable especially if patricipants are likely to object once debriefed
When’s it okay for patricipants to be mislead in a study (3)
- It’s necessary for patricipants to not know the purpose of the study in order to get realistic results
- Scientific justification
- Medical justification
What’s right to withdraw
Patricipants should be aware they can leave the study at any time and can withhold data after it’s complete
What’s protection from harm
Researchers have a responsibility to protect patricipants from physical and mental harm during the investigation- shouldn’t leave in worse state
What’s confidentiality and anonymity
• Patricipants data shouldn’t be disclosed unless agreed in advance
• Numbers should be used instead of names in published research papers
• Confidentiality means data can be traced back to names whereas anonymous data cannot as the researchers collect no names
When’s confidential data preferable to anonymous data
If patricipants are followed up layer
What’s deception in terms of ethical guidelines
All relevant details of the study should be explained to patricipants before and after especially if deceptions been used, however a debrief doesn’t justify unethical aspects
What can revealing true aims of a study to patricipants lead to
Demand characteristics
What’s the effect of demand characteristics
Reduce the internal validity of the research
What are the alternatives to informed consent
- Presumptive consent
- Prior general consent
- Retroactive consent
What’s presumptive consent
The researcher asks a group similar to the sample and if they’d agree to take part in the research it’s assumed the patricipants also would
What’s prior general consent
Patricipants agree to a long list of general features not knowing which aspects will be part of their study
What’s retroactive consent
Researcher asks for consent after the study, if the participant doesn’t agree their data’s destroyed
If the research requires deception or risks harm what happens
An ethics committee (group of experts) conducts a cost-benefit analysis
What’s a negative of a cost benefit analysis
True value of study to society isn’t known for many years so can be difficult to accurately conduct
What should be done if deceptions used in a study (3)
• Explained in debrief and aim of study
• Check no harm was done and if it was offer counselling or other assistance
• Remind them they’re still able to withdraw data and ensure confidentiality of findings
What are the types of data (4)
- Qualitative
- Quantitative
- Primary
- Secondary
What’s quantitative data
Numerical data measures in ‘how much’, ‘how many’, ‘how long’ ect
Characteristics of quantitative data (4)
- Numerical
- Objective
- Less detailed
- More reliable
How do experimental observations use quantitative data
Behavioural categories
How do questionnaires use quantitative data
Closed questions
How do interviews use quantitative data
Structured correlation
Positives of quantitative data
- Easy to analyse
- Use do descriptive stats/ statistical tests allow for conclusions to be drawn easily
Negative of quantitative data
• Can oversimplify reality
What’s qualitative data
Data about what people think and feel that cannot be counted or quantified
Characteristics of qualitative data (4)
- Non-numerical, descriptive data
- Subjective
- More detailed
- Less reliable
How do questionnaires use qualitative data
Open questions
How do interviews use qualitative data
Unstructured interviews
What type of study is qualitative in nature
Case studys
Positives of qualitative data
- Rich detailed information about people’s experiences
- Answers aren’t restricted by previous expectations- can provide unexpected insights
Negative of qualitative data
• Complexity makes it more difficult to analyse data and draw conclusions
What’s content analysis
A method of quantifying qualitative data through the use of coding units
When’s content analysis often used
With media research
Outline the process of content analysis
• Read the information and identify coding units to categorise materials (create behavioural categories)
• Read material and count each time a category occurs
What’s primary data
Refers to the original data collected specifically towards a research aim which has not been published before
What’s secondary data
Refers to original data originally collected towards another research aim which has been published before
What does collecting data involve (6)
- Designing the study
- Gaining ethical approval
- Piloting the study
- Recruiting/testing patricipants
- Analysing data collected
- Drawing conclusions
Advantages of primary data (2)
- The researcher has control over the data
- Data has been designed to fit the aims and hypothesis of the study
Disadvantages of primary data (2)
- Very lengthy and expensive process
- Designing a study takes a lot of time along with participant recruitment and conducting and analysing data
Where can secondary data come from (4)
- Own research from a previous study
- Someone else’s research
- Government stats
- Data from hospitals/ other institutions
Where’s secondary data often found
Correlational and review studies
Advantages of secondary data (3)
- Simple and cheap to access someone else’s data
- Less time consuming
- Data’s been subjected to statistical testing so it’s significance is known
Disadvantage of secondary data
Maybe it exactly fit the needs of the study
What’s reliability
The extent to which a test or measurement produces consistent results
What does reliability refer to
The consistency of results
When are findings said to be reliable
When a study’s repeated with the same method, design and measurements
What are the two types of reliability
- Internal reliability
- External reliability
What’s internal reliability
The extent to which somethings consistent within itself
Example of internal reliability
Whether all questions on a questionnaire measure the same thing
What’s external reliability
The extent to which a test measures consistently with other tests that measure the same thing
What are the 3 methods of assessing reliability
- The split half method
- The test-retest method
- The inter-rater reliability method
What’s the split half method used for
To measure internal reliability
What test measures internal reliability
The split half method
What’s the split half method
Splits the test into two halves and has the same participant do both halves, if similar results are obtained the test has a good internal reliability
What does the test retest method measure
External relivaility
What test measures external reliability
The test retest method
What’s the test retest method
Gives the same test to the same participant on two or more occasions if similar results are obtained the tests more reliable
Disadvantage of test retest method
Used for things that remain stable over time
What does the inter rater reliability method mesure
Assesses whether different observers are viewing and rating behaviours the same
What’s the inter rated reliability method
Conducting a correlation of all the observers scores to see the degree of agreement between different raters
What does a high correlation mean in the inter rater reliability method
Indicates observers are rating and categorising behaviour consistently
How can inter rater reliability be improved
Developing clearly defined and separate categories of observational criteria
