Research Methods Flashcards
What are the aims of a study
General statements of what the researcher intends to investigate
What is a hypothesis
A statement of what the researcher believes to be true. Clearly states the relationship between variables as stated by theory
What is a theory
A collection of general principles used to explain specific observations and facts
What is a directional hypothesis
States whether changes are greater or lesser positive or negative etc
What is a non directional hypothesis
Doesn’t state the direction of a study
What is the IV
The variable manipulated by the investigator
What is the Dv
The variable that is measured by the researcher
Why are all other variables beside the IV have to remain constant in a properly run experiment
So the researcher can be confident that the cause of the effect on the DV was the IV alone
What are two levels of the IV
Control condition
Experimental condition
What is operationalisation
Clearly defining variables in terms of how they cns be measured
What is an extraneous variable
Is any variable other than the IV that may have an effect on the DV. These are often nuisances that do not vary systematically with the IV but may dampen any effect such as the lighting in the lab
Example of extraneous variables
Participant variables such as intelligence
Situational variables such as distractions
What is a confounding variable
Any variable other than the IV that has affected the DV. They do change systematically with the IV so it is hard to tell if any observations come from the IV or the CV.
What are demand characteristics
Refers to any cue from the researcher or research situation that may reveal the aim of the study
What are the two demand characteristics
‘Please-U effect’
‘Screw-U effect’
What are investigator effects
Any effect of the investigators behaviour on the research outcome
What are two sources of bias
Demand characteristics
Investigator effect
What are two ways of controlling bias
Randomisation
Standardisation
What is randomisation
Refers to the use of chance when designing investigations to control the effect of bias
What is standardisation
Using the exact same formalised procedures for all participants in a research study
What are the three types of experimental design
Independent group design
Repeated measures design
Matched pairs design
What is an independent group design
One group of people do condition A another group do condition B.
Random allocation is used to assign participants to groups like drawing names out of a hat
What is a control group
A group of participants in an independent group design who receive no treatment. They act as a comparison
Strength of independent groups design
In repeated measures, order effects can occur because the same person is tested again and may do better the second time around because they know what to expect or worse because they are tired. This acts as a CV and independent groups design avoids it
What are order effects
In repeated measures design a CV arising from the order in which conditions are presented e.g a practice effect or boredom effect
Limitation of independent groups design
More participants needed than repeated measures design. They need twice as many. This means that time spent recruiting participants makes the study more expensive
Quick strength and weakness of independent group design
/ harder to guess the aim of the study
Participant variables may act as extraneous variables as everybody is different
What is a repeated measures design
Each participant does all conditions.
Order in which paritipacsnts are tested should be varied to reduce order effects.
How does repeated measures designs order effect get dealt with
Counterbalancing - half participants do condition 1 first the other do condition 2
Strength of repeated measures design
Controls participant variables. Each person acts as their own control because the person in both conditions has the same characteristics. This controls an important EV (participant variables£
Weakness of repeated measures design
Participants may do better or worse because they are doing a similar task twice. Order effects. These act as an Ev and should be compensated through counterbalancing
Quick strength and limitation of repeated measures design
Fewer participants needed than for independent groups design.
Participants may guess the aim of the study.
What is matched pairs design
Two groups of participants which acts as a compromise between independent groups and related measures as two groups are used but they are related to other by being paired.
These pairs are put into separate groups.
Strength of matched pairs design
No problem with order effects. Used two seperste groups of participants and therefore there can be no order effects nor will participants guess the aim. Enhanced the validity
Weakness of matched pairs design
Matching pairs takes time. It is a lengthy process to test participants and match them up. This increases the time taken and the costs. Matching is also not perfect
Quick strength and weakness of matched pairs design
Participant variables are partly controlled.
More participants needed than for repeated measures design
What are the four types of experiments
Labatory
Field
Natural
Quasi
What are lab experiments
Special environment where behaviour can be investigated under controlled conditions. They control EVS making it easy to control the IV. Only by controlling EVS/CVS csn we claim they the change in the IV was due to the DV
Strength and weakness of Lab experiments
EVS can be controlled and the effect on the DVs diminished. Means the experimenter can be confident in drawing a causal conclusion (high internal validity).
Lab experiments are artificial and not like everyday life. Participants know they are being observed and may behave differently meaning behaviour cannot always be generalised beyond the research setting (low external validity)
What are field experiments
Conducted in the environment where behaviour would normally occur but means there is less control compared a lab setting. IVs still controlled but less control over Ev as a result of a natural setting
Strength and weakness of field experiments
Participants are not usually aware they are being studied. Means they are likely to behave normally. Means it can be generalised. (Good external validity)
Participants are not given the opportunity to provide informed consent because they would be aware of being studied and are not always debriefed. Possible invasion of privacy raising ethical issues
What is a natural experiment
Experimenter doesn’t manipulate the IV it is a natural one such as in the study of Romanisn orphans the children were adopted without the researcher getting involved.
DV may be assessed in natural environment or in a lab.
Strength and weakness of natural experiment
In some cases it is the only ethical way research can be done. For instance in the Romanian study it would be unfair to deliberately make children experience late adoption for an experiment.
Natural events occur very rarely. Reduced the opportunity for research limiting scope for generalising findings
What is a quasi experiment
The IV is not able to be manipulated because it is a pre-existing condition such as age or gender. Participants cannot be randomly allocated to experimental conditions.
Strength and weakness of quasi experiments
Comparisons can be made regarding behaviour. Often in quasi experiments the IV is a pre-existing difference between groups of people like autism. Means comparisons can be made between people and the different behaviours they demonstrate
Random allocation to conditions is not possible. This is due to the IV being pre-existing. Means resrcger is less sure whether the IV affected the DV because other participant variables may be responsible.
What are the different sampling methods
Opportunity Volunteers Random Systematic Stratified
What does sampling mean
Selecting participants
What is the target population
Refers to the large group of people from which the sample will be drawn
What is a representative sample
One that is typical of the target population and is not biased
What do biased samples limit
The extent to which generalisations can be made
Why are most samples inevitably bias
Certain groups, ages, genders are over/under represented
What is an opportunity sample
Made up of people who are simply most available like the ones who are nearest. You just ask them to take part
Strength and weakness of opportunity sample
Quickest method to use.
Inevitably biased - unrepresentative of target population so can’t be generalised. Research bias ( may avoid some people)
What is a volunteer sample
Ask for self selecting people to take part. Might be in a news paper ad or on a notice board asking for participants
Strength and weakness of volunteer sample
Guarantees willing participants so will be willing to put more thought into the study than someone on the street.
Likely to be biased ( volunteer bias) as volunteers are likely to have a certain ‘profile’ one who is keen and helpful limiting generalisation
What is random sampling
Equal chances of selection for every member of the target population which isn’t true for opportunity or volunteer sample. Could be chosen using the lottery method where all members of population are placed in a hat and approapraite number is drawn out. Or random number tables
Strength and weakness of random sampling
Potentially unbiased as there is no infleunce over who is chosen. In reality some decline and then the remaining sample had a kind of volunteer bias - only those willing are left.
Representative sample not guaranteed.
What is systematic sampling
System is used to make a pattern which means that every men ever if the target population has an equal chance of being selected. List of all members of population is used and everything 5th or 40th etc person is used
Strength and weakness of systematic sampling
Unbiased as it’s an objective system and can be regarded as random sample if first item is selected randomly.
Takes more time and effort than other methods as a complete list of the target population has to be made much like random sampling
What is a stratified sample
Most commonly used in large scale questionnaire research. Contains participants who are selected according to their frequency in the target population. Strata are identified such as gender and the relative percentages for these are obtained and should be reflected in the sample obtained. Not a matter of equal representation in all subgroups but equal to their frequency in the target population
Strength and weakness of stratified sampling
More representative of the target population than other methods. Means generalisation of findings can become possible.
It’s not perfect. Identified strata cannot reflect all the ways that people are different within a target population so a completely representative sample is not possible
What are four ethical issues
Informed consent
Deception
Protection from harm
Privacy and confidentiality
When do ethical issues aris e
When’s conflict exists between participants rights and the researchers need to fins valuable and meaningful findings
What is the name of a code of ethical
BPS code of ethics
What is the BPS code of ethics
Quasi-legal document produced by the British Psychological Society that instructs psychologists in the UK about what behaviour is and is not acceptable when dealing with participants
What four principles does the BPS code of ethics include
Respect - appreciate the rights of others
Competence - be aware of own limits
Responsibility - avoiding harming clients or participants
Integrity - be honest and accurate
What does informed consent involve
Making participants aware of the aims, procedures, rights including right to withdraw and also what the data will be used for.
For participants under 16 what is required
Parental consent
Why does full informed consent present a dilemma
It may reveal the study’s aims and affect participant behaviour reducing validity
What are three alternatives addressing the problem of informed consent
Presumptive consent
Prior general consent
Retrospective consent
What is presumptive consent
Without getting consent from the participant themselves a sinister group of people are asked if the study is acceptable. If group agrees the original participants is ‘presumed’
What is prior general consent
Participants give permission to take part in a number of different studies including one that will involve deception
What is retrospective consent
At end of study during debriefing participants are asked for consent having already taken part in the study
What is deception
Deliberately misleading or withholding information from participants at any stage of the investigation
How can deception occur
By omission: participants may be given some information about what they will be required to do in the study but other information is withheld
False information: participants told something different to the true aim
From the researchers point of view what is deception
Harmless and necessary to conduct a valid study
What can deception be compensated by
Adequate debriefing
What four things does debriefing involve
True aim of the study
Any details not given during the study such as other experimental conditions
What data will be used for
Given the right to withhold their information if desired - important in retrospective consent
How do participants get protects from harm
The risk to participants should be no greater than in everyday life.
From the researchers point of view what is the level of potential harm
Difficult to predict, researchers may only become aware of how participants react when they are in research situations
How to deal with the issue of protection from harm
Debriefing - participant should be reassured their behaviour was typical if they feel embarrassed or anxious.
Counselling - if participants have been subject to stress or embarrassment researcher should provide it
When is it particularly important that ethical issues are dealt with
When the participants in the investigation are children. They are vulnerable and must receive special care and make sure their participation is brief as they get tired
What does privacy refer to
A persons right to control information about themselves
What is the distinction between privacy and confidentiality
We have the right to privacy, if this is invaded confidentiality/anonymity should be respected
From researchers point of view why is it difficult to know what counts as ‘public’
Some people do not wish to be observed even in public
Ways of dealing with issues in privacy
Use presumptive consent as others regard it as acceptable
Ways of dealing confidentiality issues
Use numbers instead of names or false names and don’t share the personal data with other researchers
What is the responsibility of ethical committees
To weigh up the costs and benefits of research proposals to decide whether the study should go ahead - cost- benefit analysis
What is the cost benefit analysis
Making a decision by weighing up costs like harm to the participant against gains like value of the respect
What is a pilot study
Small scale trial run of a research designed before doing the real thing
Why are pilot studies conducted
In order to find out if certain things don’t work, so you can correct them before spending time and money on the real thing
What is a pilot study trying to do
Helps the researcher Test the procedures and refine them if necessary - it’s not about getting the right results
What is a single blind study
The participant is prevented from knowing the aims so deception is involved. This is done to prevent the participants expectations affecting the participants performance
What is a double blind study
The participant and the researcher are blind to the aims of the study. A study is conducted by someone other than the main researcher to prevent investigator effects affecting someone’s performance
What is a control group
Used in experimental studies for he purpose of setting a comparison or a base line
What do control groups help to establish
Cause and effect as if the behaviour of the experimental group is different to the control group the researcher can conclude it is the effort of the IV
What are the three types of observation
Naturalist and controlled
Covert and overt
Participant and non participant
What do observations provide psychologists with
A way of seeing or listening to what people do without having to ask them
Strength of observations
Can capture unexpected behaviour. What people say they do is quite different form what they actually do so observations don’t allow for dishonesty. They are useful as they give insight into behaviour
Limitation of observation
Risk of observer bias. When collecting data for observation it may be difficult to be objective because the researchers interpretation of the situation may be affected by expectations. Involving more than one resrcher can reduce possibility of observer bias affecting validity of results.
naturalistic vs controlled observation
Naturalistic- Takes place in the setting where the target behaviour would normally occur.
In a controlled observation there is some control of the environment and EVs.
Strength and weakness of controlled observations
Easily replicated. Repeatable due to standardised procedure. Means findings can be checked to see if they occur again.
Lower validity. Behaviour produced may be contrived and artificial as a result of the setting. Can’t be applied to everyday experience.
Strength and weakness of naturalistic observation
High external validity. Behaviour is studied in the normal context so it is more natural. Means findings can be generalised to everyday life.
Low control of EV. Makes it more difficult to judge any pattern of behaviour making it More difficult to draw conclusions
Covert vs overt observations
In covert observations the participants are unaware they are being studied but in overt observations participants are aware they are being studied and they’ve given consent
Strength and weakness of covert observation
Demand characteristics are less of a factor than in overt observations. Increases the validity of the findings.
Ethics of covert observations. People may not want their behaviour noting down. Means ethics of these studies may be questioned
Strength and weakness of overt observations
Ethically more acceptable because participants have given consent to being studied. Aware their behaviour is being studied and they also have the right to withdraw their participation.
Demand characteristics are a problem. Means behaviour is less spontaneous and bust risk than it perhaps would have even, threatening the validity of the findings.
What are participant observations
The observer is part of the group they are studying for instance a study of workers and management might be improved by having the research actually join the workforce to produce a first hand account
What is a non participant observation
The researcher remains seperate from those they are studying such as trying to observer the behaviour of school children
Strength and weakness of participant observation
Lead to useful insights. Increases validity.
Possible loss of objectivity. Researcher may come to identify too strongly with those they are studying and lose obecticity. Line between being a researcher and participant becomes blurred.
Strength and weakness of non participant observations
They are more objective as the researcher keeps an objective distance from their participant so their is less chance of them ‘going native’. More validity
Loss of insight. They are far too removed form the people and behaviour they are studying reducing validity
What are questionnaires
Pre-set list of written questions to assess thoughts/ feelings. They can be used to assess the dependant variable
Strength of questionnaire
Can be distributed to a large number of people so can gather lardge amounts of data. Can also be completed without the resrcher being involved reducing the effort making it cost-effective
Limitation of questionnaire
Responses may not always be truthful. Respondents may be keen to present themselves in a positive light and this may influence their answers. Social desirability bias.
What is social desirability bias.
Tendency for respondents to answer questions in such a way that presents them in a better light
What are interviews
Face-to-face interactions. Two types of interview: structured interview made up of a predetermined question list and unstructured interview which is a free flowing interaction whrrr the interviewee is encouraged to elaborate their answers. Semi structured interview is a mix between the two where the interviewer can ask follow up questions when needed
Example of a semi structured interview
A job interview
Strength of structured interviews
Easy to replicate. Standardised format reduced the differences between interviewers
Limitation of structure interview
Can’t elaborate. It is not possible for intervieweees to deviate form topic or elaborate leading to frustration.
Strength of unstructured interview
Greater flexibility. Interviewer can follow up points as they arrive so the interviewer can gain insight into the view of the interviewed
Limitation of unstructured interview
Analysis of data is much more difficult. A lot harder to replicate so the former means that diffferences in style and types of questions between interviewers are likely introducing bias
What are the two types of self report
Questionnaire and interview
What are the theee types of observational design
Unstructured or structure observations
Behavioural categories
Observational sampling methods
What is an unstructured observation
Resrcher writes down everything they see producing accounts that are rich in detail.
When are unstructured observations appropriate
Small- scale observations and involving few participants such as a couple
What are structured observations
Predetermined list of behaviours to quantify observations they use behavioural categories
Strength and weakness of unstructured observations
Rich and detailed data is collected giving researcher more insight into behaviour than data collected in structured way.
Risk of observer bias. they may only record behaviours that catch their eye and these may not be the most important or useful. Gives unrepresentative view of participants behaviour as a whole
Strength and weakness of structured observation
Data recording and alaydis are easier. Behavioural categories makes the data more straight forwards and systematic. Quantitative data.
Key detail may be lost. Reducing observational records to numbers may loose key information. Reduced validity of eventual findings
What are behavioural categories
Structures what is being observed. Behaviour is broken into operationalised behaviours. Important that the categories are as complete as possible and that no behaviours are left out
weaknesses of behavioural categories
May be difficult to make them clear and unambiguous. They must be measurable, self evident and not overlap. For instance the difference between smiling and grinning is difficult to discern
Another limation of behavioural categories
Danger of a ‘dustbin’ category is which many different behaviours are deposited because many behaviours go unrecorded
What are the two observational sampling methods
Time sampling
Event sampling
What is time sampling
Observations st regular intervals for example once every 15 seconds the behaviour of an individual is noted
What is event sampling
Where one participant event is focused on and the researcher ticks every time the target behaviour occurs
Strength and limitation of time sampling
Reduced the number of observations that have to be made. Because observations are made at intervals. Makes data collection more structured for resrcher.
Unrepresentative. What is not recorded during time sampling may be more crucial than what is and the resrcher may miss important interactions.
Strength and weakness of event sampling
May record infrequent behaviour. If it occurs infrequently then an observer will still pick it up and they are looking for the behaviour. Event sampling enables recording of behaviours that could be missed if time sampling was used.
Complex behaviour may be oversimplified. If event the focus is on is too complex the important details may be overlooked and go unrecorded. May affect the validity of the findings
What are the two types of questions
Closed questions
Open questions
What do closed questions produce
Quantitative data. They have a specific range of answers or have a finite set of responses
What do open questions produce
Qualitative data. Allows respondents to provide their own answers to questions expressed in words
Strength and weakness of closed questions
Answers are easier to analyse. Quantitative data is easier to analyse than qualitative and can be used to produce graphs and charts so comparisons can occur.
Respondents are restricted in their answers. Can’t express precise feelings and may force answers that aren’t representative. May be frustrating.
Strength and weakness of open questions
Respondents aren’t redirected. They can elaborate and express themselves without being restricted by categories. Greater validity.
Answers more difficult to analyse. Qualitative data has more variety in answers so more difficult to analyse. May be researcher is forced to reduce data to statistics in some way in order for meaningful conclusions to be made.
How to write a good question
Avoid jargon to avoid misunderstandings.
Avoid double barrelled questions - contains two questions in one
Avoid leading questions and use neutral alternatives to emotive words.
What are four ways to ensure the interview is good
Standardised interviews schedule
Quiet room
Establish rapport
Emphasises confidential nature of the interview
What is an interview schedule
A list of questions the interviewer needs to cover
Why should interviews be conducted in a quiet room
Increase the liklihood that the interviewer will open up
How do interview establish rapport
Beginning interviews with neutral questions to make the participant feel relaxed and comfortable
When is emphasising the confidential nature of the interview important
If the interview includes topics that may be personal or sensitive
What do correlations illustrate
The strength and direction of an association between two or more co-variables
What are correlations plotted on
A scattergram with one on the x-axis and the other the y-axis.
What are he three types of correlations
Positive
Negative
Zero
What is a positive correlation
Both sets of data increase at the same time
What is a negative correlation
One co-variable increases as the the decrease
What is a zero correlation
If there is no relationship between co-variables
What is the difference between an experiment and a correlation
In experiment: resrcher controls or manipulated the IV in order to measure the effect in the DV.
In correlation: there is no manipulation of one variable and therefore it is not possible to establish cause and effect between variables.
Strength of correlation
Relatively economical to carry out. There is no need for a controlled environment and no manipulation of variables is required. Data collected by others can be used meaning correlations are less time consuming than experiments
Limitation of correlations
Causal relationships are not demonstrated. Easy to assume that is a correlation demonstrates that one co-variable has in some way caused the change in the other variable - the media often do this. Means great care should be taken to emphasis that there may be intervening variables that explain the relationships.
What are the three types of data analysis
Kinds of data
Descriptive statistics
Graphs
What is quantitative data
Numerical that can be analysed statistically
What is qualitative data
Non-numerical
Strength and weakness of quantitative data
Easier to analyse as its numerical, you can draw graphs and calculate averages. You can see patterns easily.
Oversimplifies human behaviours. Expressing thoughts and feelings in numbers means that the individual meaning is lost.
Strength and weakness of qualitative data
Represents the complexities of behaviour.using qualitative data you can explain each of these feelings in a more detailed way rather than just sayings it’s high or low. You can also include information that is unexpected.
Less easy to analyse. Large amount of detail gained is difficult to summarise. And as soon as you do it starts losing detail. Makes it difficult to draw conclusions.
What is primary data
‘First-hand’ data for example from w questionnaire, observation etc.
What is secondary data
Already exists before the investigation and has been collected by someone other than the person who is conducting the research for example in journal articles and books
Strength and limitation of primary data
Fits the job as it is gathered from the participants themsleves. Allows researcher to design the investigation so they can extract the data they need.
Requires and effort on the part of the researcher. Conducting an experiment involved time, planning and preparation. Secondary data can be accessed in minutes
Strength and weakness of secondary data
Inexpensive. Requires minimal effort making it inexpensive. When designing a study the desired information may already exist.
Quality and accuracy of secondary may be poor. Variraiton in the quality and accuracy of secondary data as much can be inaccurate or incomplete. May lack external validity
What is meta-analysis
A particular form of secondary data that involved combining data from a large number of studies. Studies should have same research questions and methods so general conclusions can be made. Can be qualitative and quantitative data - simply discuss the conclusions of studies or perform a statistical analysis of the data may involve calculating the effect size
What is the effect size
A measure of the strength of the relationship between two variables
Strength of meta-analysis
Enabled generalisations to be made. Reviewing results of many studies means the sample size is made up of a much larger set of participants. Increases the extent to which generalisations can be made strengthening the validity
Weakness of meta-analysis
Studies may not be equivalent. Research designs in the different studies may vary considerably which means studies are not always comparable. Putting them all together to calculate the effect size may be inappropriate and so conclusions may not be valid
What is publication bias
The tendency for academic journals to publish only positive findings or findings that agree with existing theory
What is the file drawer problem
Bias created because the results of some studies are not published (filed away) for example studies with negative results
What are two descriptive statistics
Measures of central tendency
Measures of dispersion
What are the three measures of central tendency
Mean
Median
Mode
What are averages
Numbers that represent typical values of a set of data
What does the phrase measures of central tendency mean
That these measures all represent a central or average figure
What is the mean
Arithmetic average.
How do you calculate the mean
Calculated by adding up all items in a set of days and dividing by the number of items
What is the median
The middle
How to calculate the median
Arranging all data in order from largest to smallest (or smallest to largest) and then selecting the middle value. If there are two values in the middle the mean of these is calculated
What is the mode
The most common
How to calculate the mode
The model group is the value that is most frequent. For use with nominal data.
Strength and limitation of the mean
Sensitive. Includes all scores in the data set giving an inclusive impression of the numerical average than the median or the mode.
Easily distorted. One anaomaly can make the mean become unrepresentative if the data. Limtation compared to the median or the model which tend not to be distorted by extreme values
Strength and limitation of the median
Not affected by extreme scores. It only focuses on middle value. Thus the median may be more representative of the data set.
Less sensitive than the mean. Not all scores are included in the calculation. Ignores the extremes of the data which may be relevant to understanding all the data
Strength and limitation of the mode
Only measure appropriate for categorical data. When data is discrete it does not make sense to use the median or the mean. Sometimes the mode is the only appropriate measure.
There may be many modes in a data set. If all scores in a data set are different the mode is every score in the data set. It is not a useful way of describing data when there are many modes.
What are the two measures of dispersion
Range
Standard deviation
What does the term dispersed refer to
How widely data is spread out
If we wish to describe data it is good to use what
Measure of central tendency and a measure of dispersion
What is range
The difference between highest to lowest (+1)
How do you work out the range
Arrange all data in order and subtract the lowest value from the highest value and adding 1.
Why do you add 1 in the range
It’s a mathematical correction and allo s for the fact that raw scores are often rounded up or down when they are recorded within research
What is standard deviation
Measure of the average spread around the mean
How dot you calculate standard deviation
Work out for each item of data the difference between the item and the mean value for the data set. The larger the standard deviation, the more spread out the data is.
Strength and weakness of range
Easy to calculate. It’s a simple formula to apply and is much easier to calculate than the standard deviation.
Does not account for the distribution of the numbers. Does not say whether most number are closely grouped around the mean or spread out evenly. Standard deviation is better in that respect.
Strength and weakness of standard deviation
More precise than range. It includes all values within the calculation. Means it gives a much more accurate picture of the overall distribution of the numbers in a data set.
May be misleading. May hide some characteristics of the data like extreme values. Extreme values may not be revealed unlike with the range.
What are the six displays of quantitative data
Visual presentation Bar chart Histogram Line graph Scattergram Table
What is looking at visual presentation of data sometimes called
Eyeballing the data
Characteristics of bar charts
Height of each column represents the frequency of that item.
Discrete data are placed along the x axis and usually frequency in the y axis.
Characteristics of histograms
The bars touch each other which shows continuous data rather than discrete.
X-axis is made up of equal-sized intervals of a single category broken down into intervaLs.
Unlike bar charts histograms have a true zero and a logical sequence.
Y-axis is frequency.
Characteristics of linegtaphs
Frequency is again usually on one axis but this time the data along the other axis must be continuous and the plotted values are joined by a line.
Line shows how something changes in value.
Characteristics of scattergram
Used for correlations analysis. It is a scatter of dots. Each dot represents a pair of data. Data on both axes must be continuous.
Characteristics of tables
Raw scores converted to descriptive statistics such as measures of central tendency and dispersion.
Data organised in columns and rows to make interpretation easier.
What are the two types of distribution
Normal distribution
Skewed distribution
What is a normal distribution
A symmetrical spread of frequency data that forms a bell-shaped pattern. The mean, median and mode are all located at the highest peak
Describe normal distribution
Most people are located in the middle area of the curve with very few people st the extreme ends. The mean, median and more all occupy the same mid-point of the curve. The tails of the curve never touch the horizontal xaxis and therefore never reach zero. They are extremely scores and are always theoretically possible
What is a skewed distribution
Not all distributions form such a balanced symmetrical pattern. Skewed distribution is a spread of frequency data that is not symmetrical, where the data clusters to one end
What is a positive skew
One where most of the distribution is concentrated towards the left of the graph resulting in a long tail on the right
Example of a positive skew
Difcult test in which most people got low marks with only a handful of students at the higher end
The mean, median and mode in a positive skew
Mode: remains at the highest point of the peak
Median: comes next to the right
Mean: is dragged across to the right
In a positive skew why is the mean dragged to the right
Extreme scores affect the mean. High scoring candidates in a test would have the effect of pulling the mean to the right
What is a negative skew
Type of distribution in which the long tail is on the negative (left side) of the leak and most of the distribution is concentrated in the right
Mean median and mode in a negative skew
Mode: highest peak
median: in the middle
Mean: pulled to the left (due to lower scorers being in the minority)
How to calculate percentages
Something divided by total number of participants multiplied by 100
How to convert a percentage to a decimal
Remove the % sign.
Move decimal two places to the left.
How to convert a decimal to a fraction
Work out number of decimal places in the number.
If there are two decimal places divide by 100 if there are three divided by 1000. E.g 83/1000.
Simplify
How do you calculate the ratios
Calculate a part- to-whole ratio. Or a part-to-part ratio.
Simplify.
What is a significant figure
Giving a rough idea of a number by substituting zeros as place holders. For example 432,567 to two significant figures is 430,000
What is standard form
Another way of expressing very large or very small numbers
How to work out standard form
Calculate how many places you move the decimal point to the left in order to get a number less than 10 or for a small number you move the decimal to the right and it’s a minus power
What does»_space; mean
Much greater than
What is the «_space;sign
Much less than
What is the proportional sign
Infinity sign without the right side
What does a wavy equals mean
Approximately equal to
What is a significant
A statistical term indicating that the resrch findings are sufficiently strong to enable a resrcher to reject the null hypothesis under test and accept the resrch hypothesis
What is a significant difference
Sufficient enough to be considered worth of notice
To find out if the difference found is significant or occurred by chance what do we do
Use a statistical test
What is the accepted level of probability in psychology
0.05
What does a 0.05 probability mean
There is a 5% probability that the results occurred by chance so hyptoshsis can be accepted
What are statistical tests designed by
Mathematicians to work out what data value would be beyond chance
What must be compared with what to determine significance
Calculated value with a critical value
What is a calculated value
Value of a test statistic calculated for a particular data set
What is a critical value
Value that a test statistic must reach in order for the null hypothesis to be rejected
What three pieces of information is needed to determine the critical value
Desired significance level / 0.05
Number of participants in the investigation / n value
Whether the hypothesis is directional or non-directional (one tailed or two)
What is the sign test for
To analyse the differences in scores between related items e.g same participants tested twice
How to calculate the sign test
- Pairs of scores for participants arranged in a table
- Score for condition B substrates from condition A to produce sign of difference (either plus or minus)
- Total number of pluses and minus calculated
- Participants who achieved the same score in condition A and B should be disregarded and deducted from the N value
- The calcuslted value is the total of the less frequent sign
For the sign test when should the experimental hypothesis be accept
If the calculated value is less than the relevant critical value in the table
Before a peice of tearful can be published ina journal what must it be subject to
Peer review
What is peer review
Assessment of scientific work by others who are specialists in the same field to ensure resrch is high quality
What should the peers in peer reviews be
Experts in the field
Objective
Unknown to the resrcher
What are three aims of peer review
Allocate resrch funding perhaps by government-run funding organisations
Validate the quality, accuracy and relevance of resrch - done by assessing hypothesis, methods, statistical tests and conclusions
Suggest amendments or improvements -or in extreme cases conclude work is inappropriate for publication
Example of a government run funding organisation
Medical resrch council
Strength of peer review
Protects the quality of published research. The fact all elements are scrutinised by experts minimises fraudulent. Also means resrch in journals is of the highest quality. Preserves the reputation of psychology as a science and increases credibility and study of the subject as a whole
Limitation of peer review
Some reviewers may use it as a way of critiquing rival resrch. They could use their animist to criticise good rival work. Resrchers are often in competition for limited funding so may give negative appraisal for this reason
Issue with publication bias in peer review
Publication bias. The natural tendency for editors to want to publish significant ‘headline grabbing’ findings to increase circulation of their publication. Could mean that good resrch is ignored or disregarded
How does attachment resrch into the role of the fathers benefit the economy
Recent research has dressed the importance of multiple attachments. If both parents are equally capable of providing for the child this promotes more flexible working arrangements in the family. If mother is the higher earner and works longner hours the couple can share the childcare responsibilities instead of it just being the mother. Means modern parents are better equipped to maximise their income
How has the development of treatments for mental illnesses affected the economy
Absense from work costs the economy £15 billion a year and a third of all days off are caused by mental disorders such as depression. Resrch into causes and treatments have an important role to play in maintaining a healthy workforce. Means sufferers can manage their conditions effectively, return to work and make a valuable contribution to the economy
What is the correlation coefficient
Statistical tests of correlation produce a numerical value somewhere between -1 and +1. The closer to either 1 is the stronger the relationship. Closer to zero the weaker the relationship.
What does the value of +1 represent
Positive correlation
Value of -1 represents
Negative correlation
What are the positives to case studies
Detailed and in depth
Unusual and also typical cases (like rare disorders or memories of childhood)
Usually qualitative but can produce quantitative data
Longitudinal
Strength of case studies
Provide rich, detailed information. May be preferred to superficial data collected from questionnaires etc. May increase validity
Limitation of case studies
Prone to researcher bias. Subject in the final report is based on subjective selection of the researcher so subject to bias. May have negative impact
What is content analysis
Observational resection in which people are studied indirectly via the communications they had produced. Aims to summarise the communication in a systematic way so conclusions scan be drawn
What is he initial stage of content analysis
Coding
What is used in content analysis
Coding units
What do coding units do
They are meaningful units which can be used to involve counting up the number of times w particular word of order appears in the text
What does content analysis produce
Quantitative data
Example of content analysis producing qualitative data
Thematic analysis
What is thematic analysis
Inductive approach to analysis that involves identifying implicit or explicit ideas within the data. Themes will often emerge once the data has been coded which can be used with a new set of data to test the validity of the themes
Strength of content analysis
It is flexible. Can be adapted to produce both qualitative and quantitative data. Means it can be adapted to suit the aims of the research
Limitation of content analysis
May lack objectivity. Modern analysts are clear about rheir own biases affecting research but content analysis still lacks objectivity especially when more descriptive forms of thematic analysis are used. May threaten validity.
What is reliability
Reliability means consistency. Tested on one day should be the same the next.
Two ways to test reliability
Test-retest
Inter-observer
Intra-observer
What is the test-retest method
Administrating the same test or questionnaire to the same person on different occasions. If the questionnaire is reliable then he results should be the same
What is test-retest method used to assess
Questionnaires or psychological tests like IQ
What is the coefficient for a strong positive relationship
Exceeds 0.80
What should the time period be between the tests in the test retest methods
Significant enough to ensure the respondents cannot recall their answers to the questions but not so long that their attitudes or abilities have chsngedn
What is inter-observer reliability
St least two people should record observations to reduce subjectivity of data collected. Resechers should establish reliability between themselves to ensure they are making similar observations
When is inter-observer reliability used
Observational research
Done in pilot studies
What is a pilot study
A small scale trial run of the observation to check the observers are applying behavioural categories in the same way
How do observers compare observations to decide if the inter-observer reliability is right
They use a pilot study where they watch the same event or sequence of events but record their data seperatedly. This should then be correlated to assess its reliability and the correlation coefficient should exceed 0.80. Other wise the researchers need to revisit and tchange the categories
What is intra-observer reliability
If resrcher is working alone the sequence of events should be recorded and observed by the same person twice. Correlation analysed between the two sets of data
How is the reliability of questionnaires improved
Replace some of the open questions with closed, fixed choice alternatives which are less ambiguous
How is the reliability of interviews improved
Use the same interviewer each time. Train all interviewed so they don’t ask leading questions or ambiguous ones.
Which type of interview is less likely to be reliable
Unstructured
How is the reliability of experiments ensured
Maintaining strict control of many aspects of the procedure, such as the instructions that the participants receive and the conditions within which they are tested. Use standardised procedures.
How is the reliability of observations improved
Making sure the behavioural categories have been properly operationalised. They shouldn’t be overlapping either. If they aren’t operationalised or are overlapping or absent the observations may be inconsistent
What is validity
Refers to whether a psychological test, observation or experience by produces a legitsmste result.
What are the types of validity
Interval validity
External validity
Ecological validity
Temporal validity
What is internal validity
Whether the researcher has managed to measure what they intended to measure
What is external validity
The extent to which findings can be generalised beyond the research setting in which they were found
An instant of data being reliable but not valid
A broken set of scale may consistently read someone’s weight wrong. The scales are reliable but the weight that is reported is not ‘true’
What is ecological validity
A form of external validity and refers to whether findings can be generalised from one setting to another, most particularly to everyday life such as learning a lists of words that people don’t usually do
Why is temporal validity
The extent to which findings from a particular study holds true over time.
Example of a study with low temporal validity
Rates of conformity in Asch’s study was a product of a particular conformist era in recent American history
What are the two ways to assess validity
Face validity
Concurrent validity
What is face validity
Does it appear to measure what it’s supposed to measure. Achieved by simply eyeballing the measuring instrument or by passing it to an expert to check that it appears to be testing what it is meant to
What is concurrent validity
When the results obtained are very close to those obtained on another recognised and well established test.
How do you improve validity in experiments
Using a control group means the researcher is better able to assess whether changes in the DV were due to the effect of the IV.
Standardise procedures.
What does standardising procedure minimise
Impact of participant reactivity and investigator effects on the validity of the outcome. The use of single blind and double blind procedures are designed to achieve the same.
How do you improve validity in questionnaires
Incorporate a lie scale within the questions in order to assess the consistency of a respondents response and to control for the effects of social desirability bias.
Also improved by assuring respondents all data is anonymous
What is a lie scale
Set of questions in a survey to determine the extent to which the participants answers are truthful
How do you improve validity in behavioural categories
Categories have to be well-defined and operationalised to improve the validity of the data collected
How do you improve validity in qualitative data
Consider the interpretive validity of their conclusions. This can be seen through the coherence of their reporting and inclusion of quotes from participants.
Triangulation - using a number of different sources as evidence
What is interpretive validity
The match between the meaning attributed to participants behaviours and the actual participants perspectives
What is triangulation
Using a number of difference sources as evidence
What are statistical tests used to see
If the results are due to chance or not.
The outcome tells the researcher whether they should accept or reject the null hypothesis
How to choose a statistical test
Difference or correlation
Experimental design
Level of measurement
What are the two types of experimental design
Unrelated design
Related design
What is an unrelated design
Participants in each condition of an independent groups design are different, therefore unrelated
What is a related design
Repeated measures and matched pairs are referred to as related designs
What are the three levels of measurements
Nominal
Ordinal
Interval
What is nominal data
Data in the form of categories. It’s discrete
What is ordinal data
Data that has been placed in rank order
Why is ordinal data sometimes referred to as unsafe data
It lacks precision because it is based on subjective opinion rather than objective measures. That’s why raw ordinal data isn’t used it is instead ranked.
What is interval data
Based on numerical scales that include units of equal, precisely defined size. Public scales of measure to that produces data based on accepted unit of measurements
What are parametric tests
A group of inferential statistics that make certain assumptions about the parameters (characteristics) of the population form which a sample is drawn
What are the different statistical test
Chi squared Mann-Whitney Unrelated t-test Sign test Wilcoxon Related t-test Chi-squared Spearmans rho Pearsons r
What is the way to remember the statistical tests
Can My Uterus Stop With Ridiculous Cramps Stupid Pricks
If the statistical test is not significant what is accepted
The null hypothesis
What do statistical tests work on
The basis of probability rather than certainty
What is a significance level
The point st which the researcher can claim to have discovered a significant different or correlation within the data. The point the resrcher chs reject the null hypothesis and accept the alternative hypothesis
What is the usual level of significance
0.05
How do you calculate df for related t-test
N-1
How do you work out df for unrelated t-test
Na + Nb - 2
What is the formula for a chi-squared test
E = (row total x column total)/ overall total
What is the calculated value
The value of a test statistic calculated for a particular data set
What is a critical value
When testing a hypothesis, the numerical boundary between acceptance and rejection of the null hypothesis
What are he three criteria the researcher must use to know what critical value to use
One- tailed or two-tailed test.
Number of participants in the study - N or degrees of freedom.
Level of significance to use (usually 0.05).
The lower the p value, the ____
More statistically significant of result
What is a type | error
Null hypothesis is rejected and the alternative hypothesis is accepted when it should have been the other way around.
Often referred to as a false positive.
What is a type || error
When the null hypothesis is accepted but he alternative should have been because they alternative was ‘true’.
False negative.
When are we likely to make a type | error
If significant level is too lenient
When is a type || error more likely
If significance level is too stringent like 0.01
Why do psychologists favour the 5% level of significance
Best balances the risk of making a type | or type || error
What are the 6 sections of a scientific report
The abstract The introduction The method The results The discussion Referencing
What is the abstract in a scientific report
It is a short summary (150-200)that includes all he major elements: aims and hypotheses, methods, results and conclusions.